U.S.news录取率最高的美国商学院听说有个商学院录取率是100%
新批评(New
新批評（New Criticism）新批評是二、三0年付形成於英美的文學批評流派，五0年付成為美國文學批評的主流。
反對把文學當成文獻、傳記、史料，注重文學本身的價值。
認為文學不同於科學，不是實用性的。
新批評強調文學內緣的研究（意象、格律、文體），主張細讀（close reading）。
新批評常用的術語：歧義（ambiguity，含混、模稜、多義）、反諷（irory）、矛盾語法（language of paradox）〃張力（tension），肌理（texture、紋理，質地）無我（impersonality），客觀相應物（objective correlative）。
敘事觀點、敘事語氣為小說討論的重點。
提出敘事者不等於作者。
詩及小說的批評重視文字、意象、結構、肌理、語氣。
強調作品必須包含各種複雜而互相矛盾的因素，只有這樣才能‚掌握人類經驗範圍的各面‛。
反對詩作為宣傳，反對大眾語言（mass language）。
新批評三0年付興起，至五、六0年付英美學術界影響仍然很大。
但也有視野較闊的批評家，如特里林（Leonard Trilling）和威爾森（Edmund Wilson）等人對他們提出批評。
在中國方面，四0年付的袁可嘉，五、六0年付香港的葉維廉、李英豪、王敬羲，美國的夏志清，台灣的歐陽子、顏元叔，八0年付的黃維樑，都受新批評的影響。
林以亮編選的《美國文學批評選》（1961）有艾略特、布魯克斯、泰德等論文。
這種批評方法，在反對載道或泛政治的批評方面有其意義，產生於特殊的歷史斷層的處境有其必然性，對文章細讀有幫助，個別評論家有進一步發展（袁可嘉、夏志清、葉維廉），亦各有盲點，如顏元叔談古詩，黃維樑談新詩，弊端特別顯著。
新批評(New Criticism) 的起源休姆（T.E. Hulme）：〈古典主義與浪漫主義〉；意象派詩Imagism龐德（Ezra Pound）詩“In a Station of the Metro”在地鐵車站(1914)：The apparition of these faces in the crowd:Petals on a wet, black bough.受俳句(Haiku)影響、受中國詩影響。
TOD ResearchonNewUrb_省略_trictofHefeiCity_778
Research on New Urban Area Planning Based on TOD Mode: A Study on Binhu New District of Hefei CityGU Dazhi, YAN Ping1. IntroductionAs a transit-oriented urban development technique, the TOD mode targets at mak-ing intensive use of land. It is widely adopt-ed in European and some Asian cities with large populations and has achieved certain success. Since the mode was introduced into China in the 1990s, it has been mainly used in mega-cities with urban rail-transit system. In fact, considering the rapid proc-ess of urbanization and accelerated urban construction, the TOD mode also has prac-tical significance for the numerous large and medium-sized Chinese cities whose current conditions do not allow for rail-transit construction①. Most of these cities are at the crucial stage of transformation from being mono-centric to being multi-centric. This paper takes Binhu New Dis-trict of Hefei City as an example to study the TOD-based new urban area design.2. Definition of TOD and its signifi-cance for Chinese citiesThe TOD mode was put forward in the the world average level, and the country has ranked No. 2 in the world in overall en-ergy consumption③. Various problems have appeared: inefficient use of land resources, immense investment in infrastructure, rap-id increase of private automobiles, heavy tidal transportation, etc. How to promote the limitations in private car use and the improvement of land-use efficiency are the main challenges that Chinese cities are fac-ing, forming the demand to adopt a more sustainable TOD mode④. The TOD mode is significant in helping to solve these urban problems in China and lead to a sustainable urban development⑤.3. Background of Binhu New District design and practical conditions for realizing the TOD modeHefei City has developed rapidly since it was established as the capital city of An-hui Province in 1949. In the 1980s’ overall development plan of Hefei, a spatial layout of “three-blade-fan model”, setting the old town area as the center and having three new urban areas built in the east,Abstract As an urban development mode, the TOD (Transit-oriented Development) mode, targeted at energy-saving and based on the public transportation system, has positive signifi cance for solving problems like disordered urban sprawl and traffi c congestion in Chinese cities. This article takes Binhu New District of Hefei as an example to illustrate the practice of the TOD mode in the urban design of large and medium-sized cities. The paper analyzes the relationship between the new district, central urban area, and public transportation development, and puts forward that the TOD mode is suitable for Bihu New District. It states that the TOD mode can be realized through methods such as the building of a “double-rapid” comprehensive public transportation network, the setting up of decentralized-compact urban clusters and TOD communi-ties, the mixevd land-use, as well as the scenery design with complete walking space. The paper also proposes that certain fl exible methods can be adopted in order to promote the TOD mode, including making full use of bus transit when rail transit is lacked, setting up TOD communities surrounding the transport hubs, etc.Keywords TOD; new urban area; road system planning; “double-rapid” comprehensive public transportation network; decentralized-compact urban cluster1980s by Peter Calthorpe, an Americanarchitect and urban planner, in order tosolve urban problems such as serioussuburbanization, large-scale low-densityurban sprawl, overdependence on mini-cars, and a high degree of social isolationin the United States. The TOD mode hasbecome an important part of planning the-ories such as “Neo-urbanism” and “SmartGrowth”. It aims at building an urbanland-use mode that suits the public trans-portation service, i.e., an urban spatiallayout that takes the public transit corridoras the tie and the comprehensive land-useclusters surrounding the traffi c stations asthe joints. Flexible to fit different urbanconditions, the TOD mode is widely usedin practice. For example, Arlington in theUnited Sates employs the subway corridordevelopment mode, which is based on railtransit stations, while Curitiba in Braziladopts the “three-direction road develop-ment mode,” which is based on the BRT(Bus Rapid Transit)②.In the past 10 years, the urbanization levelin China has developed twice as much as6667north and southwest, was formulated forHefei City. The population in Hefei has increased by 25 times in 50 years, reach-ing 1.26 million, and the urbanization rate reached 46.3% in 2002⑥. Since the urban construction could not meet the needs of urban development, it was carried out to develop a new district to solve the problem of a shortage of urban land.Binhu New District is located in the south-east of Hefei City, with Hechao Road in the west, the Southern 2nd Ring Road in the north, South Feihe River in the east, and Chaohu Lake (the fifth largest freshwater lake in China) in the south. The new dis-trict includes all the land in Dawei Town, Yicheng Town, and Yixing Town, as well as a part of the land in Luogang Town, covering a planned land-use area of 110 km 2. With abundant humanistic and natural resources in the area, Binhu New District is designed as an ecological new urban area that integrates the functions of exhibition, commerce, residence, tourism and recrea-tion. The TOD mode was adopted in the new district planning upon the analysis of the following points (see Fig. 1).(1) Relation of functional structure be-tween Binhu New District and the central urban areaCompared with the old urban center, Bin-hu New District has a limited area and is less attractive, so it is only the evacuation location for certain functions of the cen-tral urban area and has little possibility in becoming a new central area. Therefore, Binhu New District, as the extension of the central urban area, should have a close connection and convenient transportation link with the central urban area.Fig. 1 Location of Binhu New District in Hefei City(2) Relation of spatial structure between Binhu New District and the central urban areaThe new district continues the Fan-shaped development mode of the city. Like other three-wing clusters, the new district has a strong economic connection with the central urban area and requires high-speed transit system for commuting between them.(3) Development target of rail transit in Binhu New DistrictAccording to the target of the rail transit network plan of Hefei City, the first rail transit line will be completed in 2014 and will link Binhu New District and the old central area, which will become the foun-dation for adopting the TOD mode in the development of Binhu New District. In ad-dition, according to Hefei’s present trans-portation situation, the city has entered a period of acceleration with rapid increase of urban motorization. The increased tran-sit demand caused by urbanization and motorization will lead to a waste of re-sources, deterioration of environment, and traffic congestion if there are no suitablecounter-measures. Therefore, the adop-tion of a TOD mode in the new district is a natural choice, aiming to guarantee health urban development.4. Binhu New District development strategy concerning the TOD modeThe TOD mode was proposed during the beginning of the planning stage of Binhu New District. A transit-oriented new ur-ban district will be built through the set-ting up of a “double-rapid” comprehensive public transit network, the construction of decentralized-compact urban clusters and TOD communities, mixed land-use, as well as the design of walking space with scenery (see Fig. 2). The main design strategies are as follows.4.1 Building a “double-rapid” compre-hensive public transit networkConsidering the specific transportation conditions of Binhu New District, some changes are made to the conventional TOD mode. Rail transit in the west is combined with the BRT in the east to formHefei CityFirst Ring Road Second Ring RoadOld City Center First Ring Road Second Ring Road312 National RoadHechao RoadPlanning 1 Rail Transit LineFeihe RoadBinhu New DistrictChaohu Lake68Fig. 2 Overall chnography of Binhu New District Development PlanFig. 3 Transportation analysisResidential land Roads and squares Municipal facilities Green spaceSpecially -designated land Intercity transport land Warehouse land Industrial land Public facilities WatersMiscellaneous Sub-level transport hubRail transit station BRT station Bus stop fieldComprehensive transport hub Normal bus transit networksBus rapid transit Rail transit line two transportation corridors, which will serve as the framework for constructing a sub-level bus transit network covering the entire new district (see Fig. 3).4.1.1 “Double-rapid” comprehensive public transit corridorsA transit rail line is designed to run along Hechao Road to connect the old urban area with Binhu New District, which serves as the supplement of the BRT that is designed to run along South Feihe Road. These two transit lines will form the “double-rapid” comprehensive public transit corridors and become the rapid transit line reaching the central urban area. Three comprehensive transport hubs will be set up along the BRT line and be located at the three “decentral-ized-compact” urban clusters of Yicheng Town, Dawei Town, and Yixing Town. The three urban clusters will each develop its own transportation-service areas, business areas, and residential areas around the trans-port hubs that are the centers of the town ina radial shape. Multiple sub-level transport hubs will also be set up within the three ur-ban clusters, and around these hubs, a great number of dispersed TOD communities will be constructed.4.1.2 Multiple bus transit networksThe “double-rapid” comprehensive transit corridors serve as the basis for the bus tran-sit network system. The bus transit network can be divided into basic lines and supple-mentary lines and covers the entire Binhu New District area intensively. The basic lines of the bus transit network can satisfy the transit needs for medium- and short-distance travel between the three urban clusters and the TOD communities, and are responsible for the connection and transfer between the comprehensive transport hubs along the “double-rapid” transportation corridors. The supplementary lines are de-signed to satisfy the transportation needs in the undeveloped areas, as well as the marginal regions.4.1.3 Intensive graded road structurePublic transportation should be given full consideration in the design of the urban road network. The single main-street structure should be changed into an in-tensive graded and multi-functional road network structure. The main streets are designed to link the three “decentralized-compact” urban clusters, and a main road network of “four horizontal and three vertical” structure has been built in the new district. The sub-level road network is designed in the area between the main streets and is mainly located inside each urban cluster in order to divide the different TOD communities. Branch roads are built within the TOD communities. These sub-level streets and branches are mainly used to satisfy the needs of public bus transit and pedestrian.According to the need for public bus lines, the distance between sub-level streets in the city is 400 – 500 m; and the distance between branches is 200 – 300 m. As more roads share the traffic flow and improve the efficiency of bus transit, the intensive road network structure far increases the accessi-bility to the TOD communities. At the same time, the intensive road network reduces the scale of the community and the size of the streets, which makes them more comfort-able and suitable for walking.4.2 Multi-centric hierarchical spatial structureThe multi-centric structure is designed for the spatial layout of Binhu New Dis-trict. According to the plan, the three decentralized-compact urban clusters are formed naturally from north to south, from Yicheng Town to Yixing Town to Dawei Town. The comprehensive transport hubs69Fig. 5 Land-use layout within the transit station areaare designed at the eastern and westernsides of each town. Inside the urban clus-ter, sub-level bus transport hubs are built symmetrically, and the TOD communities will be constructed around these hubs. The entire Binhu New District is in a hierarchi-cal structure composed of urban clusters and TOD communities (see Fig. 4).4.2.1 Decentralized-compact urban clusters The decentralized-compact urban clusters re-gard urban public transit as the link between different multi-functional “urban organic parts” within the urban area. “Decentral-ized” refers to the “multi-centers” of the new district, and “compact” means the intensive utilization of the urban land in the area.The functional layout plan of Binhu New District adopts the idea of “decentralized-compact” urban clusters, which corre-sponds to the influence of the TOD mode. Because of the great attraction of trans-port hubs to the crowd, land development becomes more intensive when it is closer to the hubs, so that the urban construction shows an obvious phenomenon of “float-ing” to the transport hubs. Taking this feature into account, the idea of “decen-tralized-compact” urban clusters requires that the development intensity follows the principle of decreasing from the central area to the outer space, where the height of the buildings should also be lower the further they are from the hub.The three decentralized-compact urban clusters play different roles in the whole area of Binhu New District: the function of Yixing Town is exhibition; that of Yicheng Town is business; and Dawei Town is tour-ism. The development of these urban clus-ters takes the comprehensive transport hubof the “double-rapid” transit corridors as the center and reduces the developmental inten-sity when it is farther from the hub. A large ecological corridor is reserved between the urban clusters as valuable space for future development. Multiple TOD communities are constructed inside the urban clusters around the sub-level transport hubs, which integrate the functions of working, living, and recreation and form multiple mixed-use spatial units.The layout of the decentralized-compact urban cluster satis fi es the living and work-ing needs of the residents inside the TOD communities, and most of the business activities can take place inside the urban clusters so that the amount of commuting is reduced; at the same time, the design of urban clusters and the TOD communities all take the transit hubs as the centers of development, which improves the conven-ience of commuting and the efficiency of public transit. The layout of the decentral-ized-compact urban clusters based on theTOD communities is well integrated with the TOD development mode of the city.4.2.2 TOD communityThe TOD community is an urban spatial unit that connects directly with the residents. It is the main practical subject of mixed land development, transit-oriented development, and pedestrian system design. The TOD community is constructed around the sub-level transport hubs that are scattered sym-metrically in the urban cluster. The central business area with mixed commercial facili-ties and working facilities is built around the transport hubs, while the outer space is built for residential use. The development intensity is decreased from the transport hub to the peripheral area. The open space in the farthest outer ring serves as the boundary of the communities (see Fig. 5). The main plan-ning points are:① Structure: though public transit serves as the tie, it is more convenient for the residents to adopt the “sub-level tran-sit-hub-oriented” development modeFig. 4 Spatial structure Commerce and finance Dispersion-compact urban bodyGreen spaceMixed development area Municipal utilities Community support facilities Administrative office Low density residential area Industrial landMedium-high density residen-tial areaTOD community based on the compre-hensive transport hub TOD commu-nity based on the sub-level transport hub Development inten-sity: strong – weak70rather than the “comprehensive transit-hub-oriented” development mode of the BRT lines.② Scale: the plan of the new district gives up the layout of “large” residences, but rather adopts the traditional neighbor-hood pattern and reticular roads frame-work, and decides on the community scale of every block based on the radius of walking distance. Kindergartens, schools, af fi liated business facilities, and municipal facilities are scattered in the community by different service radii.③ Layout: every community is a multi-functional spatial unit whose center is the sub-level transport hub, and the central area is constructed around the hub. Living facilities such as shopping stores, supermarkets, and restaurants, as well as certain office areas, are lo-cated in the central area. Outside the center is the residential area in high density with walkable size. The outer ring is public open space that includes parks, green land, and squares.Fig. 6 Development of TOD CommunityRail transit lineBus transitTOD community based on the sub-level transport hubTOD community based on the sub-level transport hubTOD community based on the comprehensive transport hubBus rapid transitCapacity ratioVariation curve of capacity ratioCentral areaTOD community based on the comprehensive transport hubCentral functional areaOffice working area Medium-high density residential areaLow density residential areaOpen spaceOpen spaceOpen spaceMedium density residential area Low density residential areaPublic facility and working area Central functional area Office working areaMedium-high density residential area Low density residential areaMedium-high density development area Medium density development areaLow density development areaother cities and considering the present land-use situation in Hefei, the capacity of the comprehensive functional areas of busi-ness, recreation and office, which are close to the comprehensive transport hubs in the urban body, should be controlled at around 4 FAR (Floor Area Ratio), the capacity of multi-functional areas, which are close to sub-level transit hubs in the TOD commu-nities, should be around 3 FAR, and the ca-pacity of residential area should be around 2 FAR (see Fig. 6).4.3.3 Mixed-functional development of trans-port hubsIn order to encourage residents to take public transit, the seamless connection between living space, pedestrian system, and stations should be realized ⑧. Multiple seamless connection tactics are presented in the design through the mixed use of the transit hubs and business services: the station and the shopping facilities form a merging building lot (which can be as deep as three floors underground), where4.3 Mixed and high-intensity land devel-opment4.3.1 Mixed development of landOn the urban cluster level, the surrounding area of the comprehensive transport hub has the highest intensity of development in the whole Binhu New District, with high density and mixed development. The devel-opment intensity is reduced from the center to the peripheral area and the capacity is reduced by the same ratio. Inside the clus-ter, the sub-level transport hub is the center of development, around which the central business area with business facilities and of fi ce facilities are located.4.3.2 High-intensity development of landHigh-intensity development is an important feature of the TOD mode. Overseas prac-tice shows that the suitable residence densi-ty is 25 – 45 houses/acre (about 0.62 – 1.11 houses/100 m 2), and the suitable working post rate is 50/acre (about 1.24/100 m 2)⑦. But this is far from enough to cater to Chi-na’s urban development. Compared with71elevators and lifters are used to connectdifferent floors; more living facilities are set up within walkable distance from the station; the station hall and bus transfer station are built on the same fl oor, etc.4.4 Pedestrian system and public space design associated with the landscape4.4.1 Landscape system constructionNear Chaohu Lake, the new district contains the mother river of Hefei – South Feihe River – fl owing through from north to south, as well as Shiwuli River, Tangxi River, etc., and they are precious urban ecologi-cal corridors in the city. The two ecological landscape belts – the southern Feihe River landscape belt and the scenic belt around Chaohu Lake – are shaped by combining the shorelines of Chaohu Lake and southern Feihe River. The landscape belts circle the whole new district and become its “green lung”, which is mainly for tourism and rec-reational activities in Binhu New District.4.4.2 Public space designThe large area of green land between the urban complexes is preserved for an open city park. The extensive wetlands, tidal flats, and ecological farmlands constitute the pastoral and ecological landscape. The reserved green land between the TOD communities is designed for small parks, ecological controlling green land, and nursery production bases. Associated with the pedestrian system planning in each community, the public space strengthens the mutual exchange and infiltration be-tween the different functional spaces like surrounding commercial space, residential space, and cultural recreational space, etc. 4.4.3 Well-designed pedestrian systemThe TOD community provides good walk-ing and cycling access for residents. All of the streets have clear and convenient walk-ways by which the key business districts are well connected with transit facilities. All the intersections are marked with zebra cross-ings without pedestrian tunnels and over-passes. A complete bikeway system is built in or between the TOD Communities. It is situated at the origin-destination of impor-tance, such as key business districts, transit sites, employment service centers, schools and other public facilities. Within all the blocks, vehicles can go directly into the un-derground parking lots from the entrance of each block to reduce the interference to resi-dents’ walk. Various buildings should have the possible shortest distance to the transit stations, and main commercial facilities and public transportation facilities should be well combined.5. Exploring the planning model of new district based on the TOD modeThis planning exemplifies the develop-ment of Binhu New District in Hefei and studies some of the applied strategies in the development of new districts in large and medium-sized cities. The article points out that the planning of large and medium-sized cities can employ lots of flexible methods for the implementation of the TOD mode, such as making use of all kinds of public transportation to guide the urban development besides rail transit, and building TOD communities around the transport hubs so as to attain public transportation-oriented development.5.1 Making use of all kinds of public transportation to guide the urban devel-opmentThe theory and practice of the TOD modeis comparatively mature and ideal now. However, for the majority of Chinese cit-ies amid the rapid development, it is not possible and realistic to implement a TOD mode mainly based on rail transportation in a short-term period. Therefore, Chinese cities should fully adopt the other kind of transportation approaches, such as BRT, to realize the TOD mode. The successful practice of Curitiba BRT proves that they can serve such purposes as the rail trans-portation does to guide the land develop-ment along the way.Like rail transportation, BRT acting as transit-oriented service bears three traits, which are medium and high transporta-tion volume, limited stations, and per-manently basic facilities such as special highways and stations. The BRT system offers public transportation service that departs at 5-minute intervals, which is equivalent to a train with three carriages departing at 10-minute intervals. Re-garding the passenger flow of the main corridors, the one-way transportation volume of BRT is usually over 8,000 people per hour and can surpass 50,000 people per hour, which exceeds many rail transportation systems ⑨. Compared with rail transportation, the investment in other transportation approaches such as BRT is lower, and they can provide more economi-cal and realistic measures to construct the large-capacity passenger transport cor-ridors. Therefore, they have great signifi-cance in curbing the sporadic development in some Chinese cities.5.2 Planning based on the spatial distri-bution of public transport hubsThe planning of Binhu New District is made based on the spatial distribution of72transport hubs. On the one hand, the plan based on the transportation junction helps to realize the utilization of the decen-tralized-compact model as well as urban development based on transit-oriented services. On the other hand, the plan for the transit network system is also based on the hub distribution, and therefore, the rational classi fi cation and well-distributed transportation arrangements are made in this planning.6. ConclusionIn the new district planning of the past two decades, most Chinese cities have made use of urban expressways, trunk roads, or the town development axis to link the old city with the new districts. Compared with such methods, Binhu New District of Hefei City adopts the TOD mode with the following advantages: fi rst-ly, the public transportation-dominated mode will efficiently control private-car transportation, which can help to set up a reasonable traf fi c structure for the city and relieve a series of problems involving en-ergy, the environment and land resource; secondly, the high-capacity public trans-portation can meet the requirements for commuting between the new urban areas and the central urban districts; thirdly, the public transportation-dominated mode can reduce the car traffic pressure in the downtown and thus improve the mobility of the transportation system there; fourth-ly, it also reflects social justice to some extent, since the large-capacity public trans-portation system provides a lower-price but higher-quality service for many mid- and low-income families in the city. In a word, to promote the sustainable development of the urban transportation in China and to estab-lish a resource-friendly city, the TOD mode should be encouraged in developing the new urban districts to keep the orderly develop-ment of the city.(Fund: Scientific Research Project of the Ministry of Housing and Urban-Rural Development (2010-R2-21).)Edited by LI Caige Proofread by TANG YanNotes:① ZHANG Ming and LIU Jing. The Chinese Edition of Transit-oriented Development. Urban Planning Forum . 2007, 1, p. 96.② YA NG Tao, GUO Xiucheng, ZHA NG Jian, et al. Integrated Transit System in Curitiba. U rban Transport of China . 2009, 3, p. 36.③ LIU Shilin. Report on Chinese Urbanization in 2008. Shanghai: Shanghai People’s Publishing House, 2009, p. 88.④ LU Huapu. Urban Integrated Traffic Planning with Its Research Subjects Based on TOD. Bulletin of National Science Foundation of China. 2005, 4, p. 212.⑤ PA N Haixiao and REN Chunyang. Review on “Transit Oriented Development in America: Experiences, Challenges, and Prospects”. Urban Planning Overseas . 2004, 6, p. 65.⑥ WU Yongfa, ZHOU Guoyan, and XU Zhen. Green City, Hefei: Quality and Decorated. Chinese and Overseas Architecture. 2009, 2, p. 8.⑦ LI Linglan, ZHANG Guohua, and CAO Yunli. Approaches to Land-use Planning and Adjustment Adjacent to Rail Transit Stations: A Case Study on Suzhou, Jiangsu Province. Urban Transport of China . 2007, 1, p. 33.⑧ M A O J i a n g x i n g a n d YA N X i a o p e i. A n International Study on the Mutual Relationship Between Urban Transport System and Land-use. City Planning Review . 2004, 7, p. 66.⑨ R. Cervero. Transit-supportive Development in the United States: Experiences and Prospects . Washington, D.C.: Federal Transit Administration, 1993, p. 31.References:LIU Shilin. Report on Chinese Urbanization in 2008. Shanghai: Shanghai People’s Publishing House, 2009.LU Huapu. Urban Integrated Traffic Planning with Its Research Subjects Based on TOD. Bulletin of National Science Foundation of China . 2005, 4.MAO Jiangxing and YAN Xiaopei. An International Study on the Mutual Relationship Between Urban Transport System and Land-use. City Planning Re-view . 2004, 7.P. Calthorpe and W. Fulton. The Regional City: Planning for the End of Sprawl . Washington, D.C.: Island Press, 2001.PA N Haixiao and REN Chunyang. Review on “Transit Oriented Development in America: Experi-ences, Challenges, and Prospects”. Urban Planning Overseas . 2004, 6.R. Cervero. Transit-supportive Development in the United States: Experiences and Prospects . Wash-ington, D.C.: Federal Transit Administration, 1993.YA NG Tao, GUO Xiucheng, ZHA NG Jian, et al. Integrated Transit System in Curitiba. Urban Trans-port of China . 2009, 3.YU Jie, YA NG Xiaoguang, YIN Rui, et al. Public Transport Planning of Developing City Based on TOD mode. Journal of Traffic and Transportation Engineering . 2007, 6.ZHANG Ming and LIU Jing. The Chinese Edition of Transit-oriented Development. Urban Planning Forum . 2007, 1.AuthorsGU Dazhi, Registered Planner, Lecturer, Department of Urban Planning, College of Architecture and Art, Hefei University of Technology; PhD Candidate, School of Architecture, Southeast University, Nanjing, P . R. China. Research area: urban regeneration, urban design. Email: xuxu_zz@YAN Ping, Registered Planner, Deputy Director, Hefei Urban Planning Bureau; PhD Candidate, College of Architecture and Urban Planning, Tongji University, Shanghai, P. R. China. Research area: urban transporta-tion planning.。
清新蒲公英通用模板
03
添加章节标题
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et
dolore magna aliqua
03 添加章节标题
添加标题
Lorem ipsum dolor sit amet, consectetur adipisicing elit,
sed do eiusmod tempor incididunt ut labore et dolore
magna aliqua
添加标题
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua
添加标题
Lorem ipsum dolor sit amet, consectetur
adipisicing elit, sed do eiusmod tempor
添加标题
Lorem ipsum dolor sit amet, consectetur
adipisicing elit, sed do eiusmod tempor
添加标题
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua
添加标题
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua
The_Shengsheng_Philosophy_of_the_Book_of_Changes_a
Abstract: As far as its origin goes, the concept of sheng (production) or shengsheng (production–reproduction) was inherently emphasized in ancient theories of cosmogenesis informed by the Heavenly Canopy theory. From the beginning, the three versions of the Changes reflected a binary mode of thinking. The Commentaries on the Book of Changes highlight the shengsheng philosophy which consists in roughly three points: (1) the binary opposition between Heaven (higher, yang, honorable) and Earth (lower, yin, humble) constitutes the genetic structure in which the shengsheng philosophy was conceived; (2) the concept of shengsheng involves two principles, that is, “change” and “production and completion,” which are also where its value and purpose lie; and (3) the shengsheng philosophy is characterized by continuation, connection, and mutual interaction. The shengsheng philosophy can be interpreted through the threefold meaning of yi in the Commentaries on the Book of Changes, that is, simplicity, change, and changelessness. As regards this proposition, “Qian Opens the Channel” and Zheng Xuan o ered cosmogenetic interpretations, while Kong Yingda proposed one based on the substance–function theory. According to Neo-Confucians, sheng or shengsheng is the very law (principle) of the cosmos, that is, “the mind of Heaven and Earth,” and even the substance of the cosmos itself. Keywords: shengsheng, Book of Changes, Commentaries on the Book of Changes, Kong Yingda
新探索研究生英语读写教程答案微信公众号
新探索研究生英语读写教程答案微信公众号1、It’s windy outside. _______ your jacket, Bob. [单选题] *A. Try onB. Put on(正确答案)C. Take offD. Wear2、The famous writer, _____ writings for China Daily I appreciate a lot , is invited to give a speech in our university. [单选题] *A. thatB. whose(正确答案)C. whomD. who3、His understanding made a deep impression_____the young girl. [单选题] *A.on(正确答案)B.inC.forD.with4、—Where ______ you ______ for your last winter holiday?—Paris. We had a great time. （）[单选题] *A. did; go(正确答案)B. do; goC. are; goingD. can; go5、I haven’t met him _____ the last committee meeting. [单选题] *A. forB. since(正确答案)C. atD. before6、31．That's ______ interesting football game. We are all excited. [单选题] *A．aB．an(正确答案)C．theD．/7、When you’ve finished with that book, don’t forget to put it back one the shelf, ____? [单选题] *A. do youB. don’t youC. will you(正确答案)D. won’t you8、The young man had decided to give up the chance of studying abroad, _____ surprised his parents a lot. [单选题] *A. whenB. whereC. which(正确答案)D. that9、--_______ do you have to do after school?--Do my homework, of course. [单选题] *A. What(正确答案)B. WhenC. WhereD. How10、He _______ getting up early. [单选题] *A. used toB. is used to(正确答案)C. is usedD. is used for11、If you do the same thing for a long time, you'll be tired of it. [单选题] *A. 试图B. 努力C. 厌倦(正确答案)D. 熟练12、48．—________ is your new skirt, Lingling?—Black. [单选题] * A．HowB．What colour(正确答案)C．WhichD．Why13、The Yangtze River is one of （）the in the world. [单选题] *A. longest riverB. longest rivers(正确答案)C. longer riverD. longer rivers14、_____how to do with the trouble of the computer, Tom had to ask his brother for help. [单选题] *A.Not to knowB.Not knowing(正确答案)C.Not knownD.Not know15、One effective（）of learning a foreign language is to study the language in its cultural context. [单选题] *A. approach(正确答案)B. wayC. mannerD. road16、Medicines are to be taken according to the doctor’s advice. [单选题] *A. 发放B. 提取C. 配方D. 服用(正确答案)17、The manager isn’t in at the moment. May I _______ a message? [单选题] *A. take(正确答案)B. makeC. haveD. keep18、---Where’s that report?---I brought it to you ____you were in Mr. Black’s office yesterday. [单选题] *A. ifB. when(正确答案)C. becauseD. before19、If you know the answer, _______ your hand, please. [单选题] *A. put up(正确答案)B. put downC. put onD. put in20、My brother usually _______ his room after school. But now he _______ soccer. [单选题] *A. cleans; playsB. cleaning; playingC. cleans; is playing(正确答案)D. cleans; is playing the21、Sometimes only（）10 out of 500 or more candidates succeed in passing all the tests. [单选题] *A. as many asB. as few as(正确答案)C. as much asD. as little as22、Tom is very _______. He never cleans his room. [单选题] *A. lazy(正确答案)B. activeC. shyD. healthy23、His remarks _____me that I had made the right decision. [单选题] *A.ensuredB.insuredC.assured(正确答案)D.assumed24、Be careful with the knife. You may hurt _______. [单选题] *A. himselfB. ourselvesC. myselfD. yourself(正确答案)25、There are many beautiful _______ in the wardrobe. [单选题] *A. bookB. dresses(正确答案)C. cell phoneD. grocery26、Which animal do you like _______, a cat, a dog or a bird? [单选题] *A. very muchB. best(正确答案)C. betterD. well27、Finally he had to break his promise. [单选题] *A. 计划B. 花瓶C. 习惯D. 诺言(正确答案)28、17．—When ________ they leave here?—Tomorrow morning. [单选题] *A．doB．will(正确答案)C．doesD．are29、Which do you enjoy to spend your weekend, fishing or shopping? [单选题] *China'shigh-speed railways _________ from 9,000 to 25,000 kilometers in the past fewyears.A. are growing(正确答案)B. have grownC. will growD. had grown30、61．How is online shopping changing our way? ? ? ? ? ? life? [单选题] *A．of(正确答案)B．in C．on D．for。
汉字的书写 英语作文
Writing Chinese characters is an art form that has been cherished for thousands of years in China.It is not just a means of communication but also a reflection of the culture and history of the Chinese people.The process of writing these characters involves a deep understanding of their structure,meaning,and the aesthetic principles that govern their composition.The Basics of Chinese CharactersChinese characters,or Hanzi,are logograms that represent both sounds and meanings. They are composed of strokes that follow a specific order and direction.The strokes can be simple,such as a single line,or complex,involving multiple lines that form intricate patterns.The basic strokes include horizontal lines either short or long,vertical lines,dots, and hooks.The Importance of Stroke OrderThe correct stroke order is crucial in writing Chinese characters.It not only ensures that the character is legible but also contributes to the beauty of the writing.The general rules for stroke order are as follows:1.Start from top to bottom.2.Begin with the left side before moving to the right.3.Write inside before the outside.plete the upper part before the lower part.The Structure of CharactersChinese characters can be categorized into different types based on their structure:Independent characters:These are characters that stand alone and do not have any components.Compound characters:These consist of two or more components that are combined to form a new character.The components can be independent characters themselves or parts of other characters.Radicals:These are the basic components that form the basis of many characters.They often provide a clue to the meaning or pronunciation of the character.Calligraphy and AestheticsCalligraphy is the art of writing Chinese characters with a brush and ink.It is a highlyvalued skill in Chinese culture and is considered a form of meditation and selfexpression. The aesthetics of Chinese calligraphy are based on the balance,rhythm,and harmony of the strokes.The four main styles of calligraphy are:Regular script Kaishu:This is the most common and standardized form of writing.It is characterized by its neatness and uniformity.Running script Xingshu:This style is more fluid and less formal than regular script.It is often used for informal writing and is known for its elegance and grace.Clerical script Lishu:This is an ancient form of writing that is characterized by its simplicity and strength.It was used in official documents and inscriptions.Cursive script Caoshu:This is the most expressive and freeflowing form of calligraphy. It is known for its dynamic and spontaneous nature.Learning to Write Chinese CharactersLearning to write Chinese characters requires patience and practice.It is important to start with the basics,such as mastering the basic strokes and understanding the stroke order.As one progresses,they can move on to learning compound characters and practicing calligraphy.There are many resources available for learning Chinese characters,including textbooks,online courses,and writing practice sheets.The Cultural SignificanceWriting Chinese characters is not just about mastering a language it is also about connecting with the rich cultural heritage of China.Each character carries a story and a history that reflects the wisdom and creativity of the Chinese people.By learning to write these characters,one can gain a deeper appreciation for the beauty and complexity of the Chinese language and culture.In conclusion,writing Chinese characters is a journey of discovery and expression.It is a skill that requires dedication and practice but offers a unique insight into the heart of Chinese culture.Whether one is a student learning the language or a calligrapher seeking to perfect their art,the act of writing Chinese characters is a rewarding and enriching experience.。
anuscript [Word] Correspondence
![anuscript [Word] Correspondence](https://img.taocdn.com/s3/m/6c4694daad51f01dc281f121.png)
Manuscript [Word]Astronomical image restoration using an improvedanisotropic diffusionbyShin-Min Chao and Du-Ming TsaiDepartment of Industrial Engineering and ManagementY uan-Ze University, Taiwan, R.O.C.Correspondence:Du-Ming TsaiDepartment of Industrial Engineering & ManagmentY uan-Ze University135 Y uan-Tung RoadNei-Li, Tao-Y uanTaiwan, R.O.C.Fax: +886-3- 4638907E-Mail: iedmtsai@.twAstronomical image restoration using an improvedanisotropic diffusionAbstractImages secured from an astronomical telescope usually suffer from blur and from interference that scientists refer to as “noise.” Therefore, good image restoration technique has become an important tool in astronomical observation. In this paper, we propose a modified anisotropic diffusion scheme to tackle the problem of image restoration in astronomy, especially in the case of nebula images. In such images, a mass of stars may be extremely bright but also may be spread randomly in dark space, and the shape of the nebula may therefore appear obscure. T o restore the original appearance of a nebula, noisy stars must be filtered out and the detailed structure of the nebula must be well enhanced. The classical Perona-Malik anisotropic diffusion model that only considers gradient information cannot filter out noisy stars from the nebula image. In this study, we propose a modified anisotropic diffusion model that incorporates both gradient and gray-level variance information to remove “sparking” stars of various sizes and brightness in a nebula image. Experimental results from a number of astronomical nebula images have shown that the proposed anisotropic diffusion scheme can effectively remove noisy stars and maintain the shape of nebula in this particular case.Key words: Astronomical image restoration; Nebula image; Anisotropic diffusion1. IntroductionThe field of image restoration has a long history that began in the 1950s with the space program. The objective of image restoration is to reconstruct the original image from its degraded version. The image restoration techniques are widely used in various applications such as satellite imaging (Jalobeanu et al. 2000, Bretschneider 2002, Bratsolis and Sigelle 2003), medical imaging (Rathee et al. 1992, Lee et al. 2004), astronomical imaging (Molina 1994, Molina et al. 2001), forensic science (Wen and Lee 2002) and many other poor-quality imaging. In this paper, we especially focus on the problem of image restoration in astronomy.There are a few problems regarding the image secured from an astronomical telescope, such as blur and noise. The objectives of image restoration techniques for astronomy roughly include de-blurring, removal of atmospheric visibility degradation, correction of mirror spherical aberration, image sharpening, image zooming, and optimizing display (Starck et al. 2002). Much research has been done on the field of image restoration in astronomy. Molina et al. (2001) proposed Bayesian image restoration methods and used expectation-maximization (EM) algorithms to restore noise and blur in astronomical images. Starck et al. (2003) combined both the wavelet and two newly multiscale approaches of ridgelet and curvelet transforms, to the problem of restoring an astronomical image from noisy data. Due to the high-directional sensitivity of the two multiscale representations, their method can well enhance elongated features contained in the astronomical images. Wu and Barba (1998) proposed an algorithm for the restoration of star field images by incorporating both the minimum mean square error criteria and the maximum varimax criteria.The traditional image restoration techniques for astronomy mainly focus on de-blurring and noise removal. In order to restore the object effectively, most of the methods use the Point Spread Function (PSF) to de-blend images. These methods assume that the PSF in an image has been known or can be estimated based on a priori knowledge. H owever, the PSF and useful prior knowledge are not always available in most practical situations. There are other common denoising methods such as Bayesian estimate, Maximum Likelihood (ML) and Maximum A Posteriori (MAP) to estimate an original clean image from a corrupted image that contains Gaussian noise (Mihcak et al. 1999, Moulin and Liu 1999, Nikolova 1999, Dias and Leitao 2002, Deng 2004).The main difficulties in MAP estimation are the choices of a proper prior distribution of the estimated image, and the corresponding energy function to be optimized.In this paper, we propose an anisotropic diffusion scheme to tackle the problem of image restoration in astronomy, especially the nebula images. In a nebula image, a mass of stars are bright and spread randomly in the dark space, and the shape of the nebula is obscure. It’s not easy for the astronomers to observe the nebula’s outline. Figure 1 shows an example of the Henize 70 nebula image obscured with sparking stars. For the analysis of nebulae in astronomical images, we must eliminate the small bright points in the dark background. The proposed diffusion model will effectively remove sparking stars and enhance the nebula shapes without the intervention of human experts. It does not need to estimate the PSF and MAP from astronomical images.The anisotropic diffusion was first proposed by Perona and Malik (1990) for scale-space description of images and edge detection. This approach is basically amodification of the linear diffusion (or heat equation), and the continuous anisotropic diffusion is given by)],([div ),(y x ty x t t t I c I ∇⋅=∂∂ (1) where ),(y x t I is the image at time t ;div represents the divergence operator; ),(y x t I ∇ is the gradient of the image, and t c represents the diffusion coefficient.Eq. (1) will reduce to an isotropic diffusion equation if the diffusion coefficient t c is a constant. It is then equivalent to convolving the image with a Gaussian filter. The idea of anisotropic diffusion is to adaptively choose t c in different iterations so thatintra-regions in an image become smooth while edges of inter-regions are sharply preserved. The diffusion coefficient t c is generally selected to be a nonnegativemonotonically decreasing function of gradient magnitude so that small variations of intensity such as noise or shading can be well smoothed, while edges with large intensity transition are distinctly retained.Hamza et al. (2002) described that the MAP estimator for a noiseless image u can be given by}2)({min arg ˆ20u u u F u u −+=λwhere uˆ is an estimate for u ;0u is the observed image; λ is a constant, and F is a given energy function. Y ou et al. (1996) considered the anisotropic diffusion as the gradient descent method for solving an energy minimization problem. They also showed that the Perona-Malik diffusion in eq. (1) is the gradient descent flow for aproperly selected energy function )(u F .Barash (2002) addressed the fundamental relationship between anisotropic diffusion and adaptive smoothing. He showed that an iteration of adaptive smoothing ¦¦¦¦++++++=+i j t i j t j y i x w j y i x w j y i x y x ),(),(),(),(t 1t I I (2)is an implementation of the discrete version of the anisotropic diffusion equation ifthe weight t w in eq. (2) is taken as the same of the diffusion coefficient t c ineq.(1). The anisotropic diffusion approach has become an useful tool for edge detection (Alvarez et al. 1992, Chen and Barcelos 2001), image enhancement (Sapiro and Ringach 1996, Solé and López 2001), image smoothing (Torkamani-Azar and Tait 1996, Tsuji et al. 2002), image segmentation (Niessen et al. 1997, Bakalexis et al. 2002), texture segmentation (Deng and Liu 2000), defect detection (Tsai and Chao 2005) and image restoration (Y ou and Kaveh 1999). H owever, the anisotropic diffusion approach has not been used in image restoration in astronomy. In order to restore the nebula outline obscured by sparking stars, the stars should be filtered out from the astronomical image. The classical anisotropic diffusion model of Perona and Malik (P-M model) only considers the gradient information of the image. When using the P-M model to restore a astronomical nebula image, the bright stars in a dark background will result in large magnitude of gradient, and can not be filtered out successfully. In this study, the gray-level variance information along with the gradient is added to a modified anisotropic model for image restoration. The proposed method can effectively remove the tiny sparking stars and enhance the shape of nebula in therestored image.The organization of this paper is as follows: Section 2 first overviews the anisotropic diffusion equation of Perona and Malik. The proposed anisotropic diffusion scheme that adaptively smoothes or retains gray levels by taking into account both gray-level variance and gradient for astronomical image restoration is then discussed. Section 3 presents experimental results from a number of astronomical nebula images. This paper is concluded in Section 4.2. The modified anisotropic diffusion model2.1 The Perona-Malik anisotropic diffusionLet ),(y x t I be the gray level at coordinates ),(y x of a digital image at iteration t , and ),(0y x I the original input image. The continuous anisotropicdiffusion in eq. (1) can be discretely implemented by using four nearest neighbors and the Laplacian operator (Perona and Malik 1990):¦=+∇⋅+=41)],(),([41),(),(i y x y x y x y x i t i t t 1t I c I I where ),(y x i t I ∇,i = 1, 2, 3 and 4, represent the gradients of four neighbors in the north, south, east and west directions, respectively, i.e.,),()1,(),(1y x y x y x t t t I I I −−=∇),()1,(),(2y x y x y x t t t I I I −+=∇),(),1(),(3y x y x y x t t t I I I −+=∇),(),1(),(4y x y x y x t t t I I I −−=∇),(y x i t c is the diffusion coefficient associated with ),(y x i t I ∇, and is considered as a function of the gradient ),(y x i t I ∇in the P-M model, i.e.,)),((),(i t it y x g y x I c ∇=For the sake of simplicity, ),(y x i t I ∇ is subsequently denoted by I ∇. The function )(I ∇g has to be a nonnegative monotonically decreasing function with 1)0(=g and 0)(=∇∞→∇I g lim I . The function )(I ∇g should result in low coefficient values at image edges that have large gradients, and high coefficient values within image regions that have low gradients. In the Perona-Malik anisotropic diffusion model, a possible diffusion coefficient function is given by])(1[1)(2K I I ∇+=∇g (3) where the parameter K is a constant, and must be fine-tuned for a particular application. Parameter K in the diffusion coefficient function acts as an edge strength threshold. If the K value is too large, the diffusion process will oversmooth and result in a blurred image. In contrast, if the K value is too small, the diffusion process will stop the smoothing in early iterations and yield a restored image similar to the original one.Let )φbe a flux function (Perona and Malik 1990) defined by(I∇φ(4)∇)()(g=∇II⋅I∇A large flux value indicates a strong effect on smoothness. Figures 2 and 3 depict the diffusion coefficient function and the flux function in eqs. (3) and (4), respectively. For a given K value, it can be seen from Figure 2 that the diffusion coefficient function in eq. (3) drops dramatically and approximates to zero when the gradient magnitude I∇is larger than 5K. That is, the diffusion stops as soon as I∇>5K. The maximum smoothness occurs at K1∇I, as shown in the corresponding flux=function. The classical P-M model considers only the gradient information of image for image restoration. Therefore, it can not effectively eliminate noises with large gradient or preserve target objects with low gradient in an ill-structured image.2.2 The proposed anisotropic diffusionIn this study, our objective is to restore the unclear shape of a nebula obscured by sparking stars in an astronomical image. Observing the previous nebula image in Figure 1, we find numerous bright stars spread randomly all over the image. Those tiny bright stars have very high gray values. They may embedded in the dark background of low gray values, or in front of the nebula region of high gray values. It indicates that the traditional P-M model that only takes into account the gradient information can not sufficiently remove the stars even after a long sequence of iterations. Figure 4 shows the result after 50 iterations of the diffusing process usingK8. It can be found from the P-M model with a properly selected parameter value =the figure that the stars are still retained, but the details of the nebula have been destroyed in the restored image. It is not an acceptable result for astronomers to distinguish the complexion of the nebula.Since the stars have bright intensities and scatter in the dark space in a nebula image, it infers that there are larger variances of gray levels in the neighborhood of stars, compared with those of the nebula. In order to remove sparking stars effectively in a nebula image, we incorporate the local variance information of gray levels in the diffusion model. By including the gray-level variance in the diffusion process, the diffusion coefficient function in eq. (3) is revised as])(1[1)(222σσ⋅∇+=∇0K II ,g (5)where 0K is a positive constant, and 2σ is the local variance of gray levels in a 33× neighborhood window. For a given pixel of coordinates ),(y x at iteration t ,the variance is defined by211112)),(),((91),(¦¦−=−=−++=i j ty x j y i x y x t t I I σ where ),(y x t I is the mean of gray levels in the 33× neighborhood window. If 2σ is fixed throughout the entire image, 2σ⋅0K will become a constant and the modified diffusion model with the diffusion coefficient function in eq. (5) will be equivalent to the P-M model. The modified diffusion coefficient function in eq. (5) isa function of two variables, the gradient I ∇ and the gray-level variance 2σ. The new flux function is given byI I I ∇⋅∇=∇),(),(22σσφg (6)In a nebula image, noisy stars are bright and spread randomly in the dark space. With such characteristics, stars have both a large gradient value I ∇ and a large variance value 2σ, and 2σ is more significant than I ∇ in terms of magnitudes. However, the gray levels of a nebula are changed gradually in the dark space. In the nebula regions, the gray-level variance 2σ is relatively small, and is not as significant as the gradient magnitude I ∇. Figure 5 presents the 3D curved surface of the diffusion coefficient function in eq. (5) as functions of 0K I ∇ and 2σ. Whenthe gradient magnitude I ∇ increases, the value of the diffusion coefficient function is reduced gradually as occurs in the traditional P-M model does. However, when 2σis significantly larger than 0K I ∇, the function value will dramatically increase toone and results in a strong smoothing in the diffusion process. Therefore, the stars can be effectively filtered out in the restored image. When 0K I ∇ is reversely largerthan 2σ, the value of the diffusion coefficient function will be close to zero and the smoothing operation will not be carried out in the diffusion process. Therefore, the shape of nebula can be completely preserved in the restored image. Figure 6 illustrates the 3D curved surface of the flux function in eq. (6). It represents the degree of diffusion effect as a function of 0K I ∇ and 2σ. When both I ∇ and 2σ increasesimultaneously, the value of flux function will dramatically increase. That is, the high variance and high gradient strengthen the diffusion process since a large flux value indicates a strong effect on smoothness. In a nebula image, the bright stars havehigher variance and higher gradient than those of the nebula. The noisy stars can be effectively eliminated and the shape of nebula can be well enhanced by incorporating both gray-level variance and gradient in the diffusion model.In the P-M model, the degree of diffusion is based on the gradient value with respect to a constant parameter K , as seen in eq. (3). In the P-M model, the value of parameter K is fixed throughout the entire image in every iteration. W e can describe the modified diffusion model that involves two variables of gradient and local variance from the aspect of adaptive K . In the proposed diffusion coefficient function of eq. (5), 2σ⋅0K can be considered as an adaptive version of K in the P-M model (eq. (3)). Each individual pixel in the modified diffusion model has its own K value, which is determined by the magnitude of its local variation. Theadaptive K function for a pixel at coordinates ),(y x is definedby ),(),(2y x y x σ⋅=0K K (7)where ),(2y x σ is the variance defined in a small neighborhood of ),(y x . ),(y x K is proportional to the variance ),(2y x σ. When ),(y x K is significantly larger than the gradient I ∇, the value of diffusion coefficient function will also be large and the modified model gives a strong smoothing. A noisy star has a large variance and,therefore, it results in a large K value for strong smoothing. When I ∇ isreversely larger than ),(y x K , the function value will be small and the modified diffusion model stops the smoothing. This prevents the blurring of nebular edges.Figure 7 shows the curves of diffusion coefficients for the P-M model and themodified diffusion model under a given I ∇. For a given value of I ∇, the diffusion coefficient t c of the P-M model will be the same since a constant K is used.However, the value of t c in the modified diffusion model will vary according to thelocal variance since the parameter K is adaptively determined by 2σ⋅0K . The value of t c increases as the value of 2σ increases. In our study, it can be inferredthat the variance of a nebula area may locate in region A in Figure 7 because the gray levels around the nebula are changed gradually and smoothly. Conversely, the variance of noisy stars may locate in region B in Figure 7 because the stars have bright intensities in the dark background. The P-M model cannot find a suitably fixed value for the parameter K that can simultaneously remove noisy stars and preserve the shape of a nebula in an image. A small fixed K value disables the diffusion process and cannot remove the noisy stars effectively. A large fixed K value will oversmooth both nebula and stars. In a nebula area (region A), the modified diffusion model with a small adaptive K provides little smoothing effect and preserves the details of the nebula. In a noisy star area (region B), the modified diffusion model with a large adaptive K yields a stronger smoothing effect and well eliminates the noisy stars. The adaptive parameter K allows the modified diffusion model to eliminate noisy stars and retain a nebular shape at the same time without conflict.Since the parameter 0K must be fine-tuned for a particular application, the following experiments are performed to find out the suitable value of 0K for the restoration of nebula images. Figure 8(a)-(f) shows the restored results of the sample image in Figure 1 for various values of 0K =5, 1, 0.5, 0.1, 0.05 and 0.01, respectively. The number of iterations is set to 10 for all tests. When 0K is overly large, the resulting images are severely blurred. Not only the stars are totally filtered out, butalso the detailed features of the nebula is lost, as seen in Figures 8(a)-(c).In contrast, K is overly small, the restored result in Figure 8(f) shows that the diffusion whenprocess cannot effectively remove the noisy stars and the filtered image is similar to the original one.Figures 8(d) and (e) both have a good effect on eliminating the noisy stars. The shape of nebula restored in Figure 8(e) is clearer than the one shown in Figure 8(d). By considering the effective removal of noisy stars and preservation ofK in eq. (5) is 0.05 for nebula details in the restored image, the suitable value ofK in eq. (5) the application of nebula image restoration. Note that the value ofdepends on the intensity range of the input image.3. Experimental ResultsIn this section, we present experimental results from a number of astronomical nebula images. All the nebula images for testing were obtained from the Anglo-Australian Observatory (.au/). The algorithms were implemented on a Pentinum 4, 3G H z personal computer using the Visual Basic200×pixels wide with 8-bit gray levels. The value language. The images were 200K in the diffusion coefficient function of eq. (5) was set to a fixed of parametervalue of 0.05, and the number of iterations equals to 10 for all test images in the200× image was following experiments. Computation time of 10 iterations on a 2000.2 seconds.The test images in Figures 9(a1)-(a3) present three types of nebula images obscured with sparking stars. The restoration results from the traditional P-M diffusion model that considers only the gradient information are shown in Figures 9(b1)-(b3). It can be found from the figure that the bright stars can not be removedeffectively and the detailed structures of the nebulae are destroyed. Figures 9(c1)-(c3) show the restored results from the proposed diffusion model that takes into account both gradient and gray-level variance. From the figure, it appears that the noisy stars are effectively filtered out and the original appearance of the nebula is very well preserved.Figures 10(a1)-(a3) present three additional nebula images that contain more noisy stars and make the nebulae hardly visible. The structures and appearances of these three nebula images are more obscure, compared with those in Figure 9. Figures 10(b1)-(b3) show the results from the traditional P-M diffusion model. As with those shown in Figures 9(b1)-(b3), the restored images are not satisfactory with the P-M model. Figures 10(c1)-(c3) present the restored images from the proposed diffusion model. It can be found that the proposed method can reliably eliminate all sizes of bright stars and keep the shape of the nebula visible to observers.In order to further verify the performance of the proposed method, the commonly denoising techniques including Gaussian smoothing, median filtering and Fourier transform were used for comparison. Figures 11(a)-14(a) show the restored results from the proposed method that incorporates both gradient and local variance information, and Figures 11(b)-14(b) present the restored images by using the Gaussian smoothing, in which the noisy stars cannot be sufficiently removed. Figures 11(c)-14(c) show the restored images from the median filter. It can be observed that some smaller stars can be eliminated, but larger bright stars still remain in the filtered images. The Fourier transform blurs the whole image including the nebula and stars, as seen in Figures 11(d)-14(d).In the experiments, two locally adaptive denoising algorithms that use local variances to estimate the desired intensities in the original image are also evaluated. The two adaptive donoising methods selected are the Wiener filtering (Lim 1990) and a wavelet-based bivariate shrinkage (Sendur and Selesnick 2002). The Wiener filtering evaluated in this paper uses local mean and local variance information to filter out noise (MA TLAB 2004). The Wiener filtering method performs little smoothing when the local variance is small, and does the opposite when the variance is large. The bivariate shrinkage function uses locally adaptive estimated variances in the wavelet domain to estimate the original image. Figures 11(e)-14(e) and 11(f)-14(f) show the restored images from the Wiener filtering and the wavelet-based bivariate shrinkage method (wavelet shrinkage), respectively. It can be observed from the figures that some larger bright stars are retained in the filtered images for both adaptive denoising methods. The results reveal that solely considering the local variance cannot effectively eliminate noisy stars. The commonly used smoothing methods and the two locally adaptive denoising methods evaluated in the experiments either fail to remove noisy stars of large size or blur the details of a nebula. The results indicate that the proposed diffusion method that considers both gradient and variance of gray levels is well suited for the restoration of a nebula in an astronomical image.4. ConclusionsIn this paper we have proposed a modified anisotropic diffusion scheme for astronomical image restoration. The astronomical images targeted in this study are unclear nebulae obscured by stars. The noisy stars in a nebula image have different sizes and brightness, and are spread randomly. Because a large number of starsobscure the nebula all over the image, astronomers can not easily observe the outline of the nebula. In order to restore the original appearance of a nebula, the stars must be filtered out and the detailed structure of the nebula must be well enhanced. The traditional P-M diffusion model only considers the gradient information of gray levels and, therefore, can not effectively eliminate the noisy stars. Since the stars have higher intensities in the dark space, the stars involve higher variance of gray levels in the image. The proposed diffusion method incorporates the variance information of gray levels into the traditional diffusion coefficient function to filter out noisy stars in the image. Experimental results have shown that the proposed anisotropic diffusion scheme can effectively remove noisy stars, and yet maintain sharp edges of a nebula in the astronomical image. It can help the astronomers observe the nebula more easily from the restored image. It is believed that the proposed method can be extended for removal of impulse noise in general, and it is currently under investigation.ReferencesAlvarez, L., Lions, P.L., Morel, J.M., 1992. Image selective smoothing and edge detection by nonlinear diffusion. SIAM Journal on Numerical Analysis Archive 29, 845-866.Bakalexis, S.A., Boutalis, Y.S., Mertzios, B.G., 2002. Edge detection and image segmentation based on nonlinear anisotropic diffusion. In: IEEE International Conference on Digital Signal Processing 2, Santorini, Greece, 1203-1206.Barash, D., 2002. A fundamental relationship between bilateral filtering, adaptive smoothing, and the nonlinear diffusion equation. IEEE Transactions Pattern Analysis and Machine Intelligence 24, 844-847.Bratsolis, E., Sigelle, M., 2003. Fast SAR image restoration, segmentation, and detection of high-reflectance regions. IEEE Transactions Geoscience and Remote Sensing 41, 2890-2899.Bretschneider, T., 2002. On the deconvolution of satellite imagery. In: IEEE International Geoscience and Remote Sensing Symposium 4, Toronto, Canada, 2450-2452.Chen, Y., Barcelos, C.A.Z., 2001. Smoothing and edge detection by time-varying coupled nonlinear diffusion equations. Computer Vision and Image Understanding 82, 85-100.Deng, G., 2004. EM algorithms for robust signal filtering and prediction. European Signal Processing Conference, Vienna, Austria.Deng, H., Liu, J., 2000. Unsupervised segmentation of textured images using anisotropic diffusion with annealing function. In: International Symposium on Multimedia Information Processing, University of Sydney, Australia, 62-67.Dias, J.M.B., Leitao, J.M.N., 2002. The ZʌM algorithm: a method for interferometric image reconstruction in SAR/SAS. IEEE Transactions Image Processing 11, 408- 422.H amza, A.B., Krim, H., Unal, G.B., 2002. Unifying probabilistic and variationalestimation. IEEE Signal Processing Magazine, 19, 37-47.Jalobeanu, A., Blanc-Feraud, L., Zerubia, J., 2000. Satellite image deconvolution using complex wavelet packets. In: International Conference on Image Processing 3, V ancouver, BC, Canada, 809-812.Lee, K.J., Papadakis, N.G., Barber, D.C., Wilkinson, I.D., Griffiths, P.D., Paley,。
屠呦呦英文介绍
• Tu started her malaria research in China when the Cultural Revolution was in progress. In early 1969, Tu was appointed as the head of the project, named Project 523 research group at her institute. She and her colleagues experimented with 380 extracts in 2,000 candidate recipes before they finally succeeded.
• Effective combination of TCM and WM. There are enormous treasure in TCM, so it can play a more important role in modern medicine. In addition, a simple application of Western medical standards to TCM won't work, Chinese medicine development needs more open inclusive of Chinese and Western to make breakthroughs.
• Improving the innovation mechanism. China should focus on cultivating good academic research environment and scientific research mechanism to encourage more scientists. China spends more than 1 billion yuan ($156 million) annually on TCM-related research, and that investment should be further strengthened.
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
U.S.news录取率最高的美国商学院听说有个商学院录取率是
100%
美国一哈佛商学院为首的的商科教育处在全球领先水平，顶级商学院的毕业生就业前景良好，专业知识水平较高，因此就任职位也比较理想。
商科也是中国留学生比较热衷选择的专业，但是美国商学院的录取要求都比较高，就算书案例进入商学院学习，每年的高淘汰率也不一定能够让你顺利毕业，所以商学院毕业的“精英们”备受企业追捧。
今天点课台君给同学们介绍一下U.S.news评选出的MBA专业录取率最高的10所美国商学院。
学校（州）
中文名
全日制学生申请数
全日制学生录取数
全日制学生录取率
University of South Florida
南佛罗里达大学
21
21
100%
Northern Arizona University (Franke)
北亚利桑那大学
52
50
96.20%
Belmont University (Massey) (TN) 贝尔蒙特大学
37
34
92%
Missouri University of Science
& Technology
密苏里科技大学
145
123
84.80%
Appalachian State University (Walker) (NC)
阿帕拉契州立大学
32
27
84%
Coastal Carolina University (SC) 卡罗来纳海岸大学
69
58
84.10%
Oklahoma State University (Spears)
俄克拉荷马州立大学静水分校
53
44
83%
Tulane University (Freeman) (LA)
杜兰大学
129
101
78.30%
University of Colorado—Boulder (Leeds) 科罗拉多大学波尔得分校
259
194
75%
Clarkson University (NY)
克拉克森大学
144
105
72.90%
南佛罗里达大学
南佛罗里达大学是63所美国国家一流的公立研究大学中的一所，在2011年被评为全美一级大学。
入学要求：本科 TOEFL要求：80 IELTS要求：6.5 GPA要求：3
研究生 TOEFL要求：80 IELTS要求：6.5 GPA要求：3
学费：学费 17000-20000美元/年
生活费 8000-10000美元/年
总费用 30000美元/年
商学院专业设置：会计、广告、商务行政与管理、商务管理经济学、金融-综合、商业-综合、国际商务管理、管理信息系统、营销管理
北亚利桑那大学
建立于1899年的州立公立大学，《企业家》杂志将学校的弗兰克商学院列为美国毕业生满意度最高的15所大学之一，该学院的MBA项目曾被《普林斯顿评论》评为全美最优秀的商学硕士项目之一。
自2006年起，弗兰克商学院每年十月都会收到“股神”巴菲特的邀请，派出由20名优秀学生组成的访问团前往奥马哈参加巴菲特问答会并与其共进午餐。
弗兰克商学院是全美唯一连续五年收到邀请的学院。
入学要求：学士申请高中毕业，GPA3.25,TOEFL525(机考195)分，雅思6.0以上，没有语言成绩可以先读ESL。
硕士申请者要有学士学位，GPA3.25，TOEFL550，雅思7以上。
MBA需要GMAT。
GRE根据专业而定。
有些专业需更高的TOEFL成绩，MBA、应用语言学(PHD)和物理治疗需600以上，公共科学580以上，数学、对统计学和TESL需570以上。
无ESL。
学费：本科:学费$13026，生活费$12974;经济资助:第1学年$8500，第2-第4学年$5500，
夏季入学的学生进入秋季之后才能得到8b部分。
硕士:学费$11422
生活费$15382经济资助:第1年$3000，ESL:6月和7月入学(5周):$2406，1、8月(16周):$9760。
经济资助:纯ESL的学生有每学期$1000。
翻页查看其他商学院的介绍和入学要求???。