| 摘要: |
| 一直以来，如何认知乡村景观空间，如何揭示与表达乡村景观的地方性特征，是开展乡村空间研究与规划设计实践的基本命题。景观单元的识别与划分是生态智慧视角下的乡村景观空间解构与重组的科学途径。利用人工智能深度学习技术中的U-Net与BDCN模型对南京市牌坊社区村域范围内的景观要素与景观单元边界进行图像识别，完成乡村社区尺度下的基本空间单元划分。选取乡村景观特征相关指标，使用空间聚类的方法完成由基本空间单元到复合空间单元的聚合与重组。通过评价复合空间单元的景观绩效认知牌坊社区的空间生态智慧，以此作为乡村景观塑造的有效引导。 |
| 关键词: 景观空间单元 空间解构 深度学习 空间聚类 景观绩效 乡村景观 生态智慧 |
| DOI：10.14085/j.fjyl.2022.07.0030.07 |
| 分类号:TU982.29 |
| 基金项目:国家重点研发计划（编号 2019YFD1100405） |
|
| Spatial Ecological Wisdom for Identification and Cognition of Rural Landscape Units: A Case Study of Nanjing Paifang Rural Community |
|
CHEN Zhaofang, WANG Yuncai
|
|
Tongji University
|
| Abstract: |
| How to cognize rural landscape space and reveal and express the local characteristics of rural landscape has long been an elementary proposition in rural spatial research, planning and design. The identification and division of spatial units is a scientific way to deconstruct and restructure rural landscape space from the perspective of ecological wisdom. By virtue of U-Net and BDCN models involved in AI deep learning technology, this research implements image identification of landscape elements and landscape unit boundaries within the scope of Nanjing Paifang rural community, in an effort to complete the division of basic spatial units therein at the scale of rural community. Additionally, by selecting the character indexes of rural landscape and using the method of spatial clustering, the research completes the transformation from basic spatial units to composite spatial units by means of aggregation and reorganization. The research cognizes the spatial ecological wisdom of the Paifang rural community by measuring the landscape performance of composite spatial units therein, which can be used as a practical guide for the construction of rural landscape. |
| Key words: landscape space units spatial deconstruction deep learning spatial clustering landscape performance rural landscape ecological wisdom |
