Scholarlink

<p>With the acceleration of urbanization, the evolution of urban landscape space has an important impact on the sustainable development of cities. Based on remote sensing technology and machine learning method, this paper discusses the evolution law of urban landscape space and the optimization of sustainable development path. By obtaining the remote sensing image data of a city from 2000 to 2020, the deep learning algorithm is used to identify and classify the changes of urban landscape space, and the dynamic changes of urban land use types are analyzed. The study found that in the past 20&#xa0;years, the city's urbanization level has improved significantly, with urban construction land increasing from 150&#xa0;square&#xa0;kilometers in 2000–280&#xa0;square&#xa0;kilometers in 2020, an increase of 86.7%. Statistical analysis using a paired t-test showed that the changes in agricultural land and water area are statistically significant (<i>p</i> &lt; 0.05), with agricultural land decreasing by 35% and water area decreasing by 15%. Regression modeling indicates that urbanization (X) significantly predicts the reduction in green space and water area (Y), with a correlation coefficient of −&#xa0;0.78, while commercial and residential land areas increased significantly, accounting for 60% of the total urban area from 40% in 2000 to 2020. This study utilizes machine learning models, including Convolutional Neural Networks (CNNs), Support Vector Machine (SVM), and Random Forest (RF), to predict urban landscape spatial evolution. CNNs are employed for image recognition and segmentation due to their ability to extract high-level spatial features from remote sensing data, while SVM and RF are used for land classification tasks, providing a robust comparison in terms of accuracy and performance. This study uses historical data from 2000 to 2020 to analyze the urban landscape changes. For future projections, we simulate the evolution of urban landscapes from 2020 to 2040, assuming the continuation of the current development mode. These projections indicate that urban construction land will expand further, potentially increasing ecological pressures. Therefore, combining the spatial data characteristics and ecological carrying capacity of urban landscape, this paper puts forward two sustainable development paths: one is to optimize the land use layout and increase the green space and water area; The second is to promote the construction of low-carbon buildings and green infrastructure. By simulating and analyzing the spatial evolution of urban landscape under different paths, optimizing paths can effectively reduce the pressure of ecological environment and improve the sustainable development ability of cities. The data analysis shows that the optimized path 1 can increase the urban green space and water area by more than 30% and reduce the proportion of construction land by 20%, thus alleviating the pressure on the ecological environment; The second path is to reduce urban carbon emissions by about 15% by promoting green building construction. To sum up, this study provides a scientific basis for the sustainable development path of urban landscape space, and provides data support and decision-making reference for policy makers.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on spatial evolution and sustainable development path optimization of urban landscape based on remote sensing and machine learning

  • 期刊论文

  • 发表日期：

    2025年11月12日

查看全文
  • Limin Song

摘要

With the acceleration of urbanization, the evolution of urban landscape space has an important impact on the sustainable development of cities. Based on remote sensing technology and machine learning method, this paper discusses the evolution law of urban landscape space and the optimization of sustainable development path. By obtaining the remote sensing image data of a city from 2000 to 2020, the deep learning algorithm is used to identify and classify the changes of urban landscape space, and the dynamic changes of urban land use types are analyzed. The study found that in the past 20 years, the city's urbanization level has improved significantly, with urban construction land increasing from 150 square kilometers in 2000–280 square kilometers in 2020, an increase of 86.7%. Statistical analysis using a paired t-test showed that the changes in agricultural land and water area are statistically significant (p < 0.05), with agricultural land decreasing by 35% and water area decreasing by 15%. Regression modeling indicates that urbanization (X) significantly predicts the reduction in green space and water area (Y), with a correlation coefficient of − 0.78, while commercial and residential land areas increased significantly, accounting for 60% of the total urban area from 40% in 2000 to 2020. This study utilizes machine learning models, including Convolutional Neural Networks (CNNs), Support Vector Machine (SVM), and Random Forest (RF), to predict urban landscape spatial evolution. CNNs are employed for image recognition and segmentation due to their ability to extract high-level spatial features from remote sensing data, while SVM and RF are used for land classification tasks, providing a robust comparison in terms of accuracy and performance. This study uses historical data from 2000 to 2020 to analyze the urban landscape changes. For future projections, we simulate the evolution of urban landscapes from 2020 to 2040, assuming the continuation of the current development mode. These projections indicate that urban construction land will expand further, potentially increasing ecological pressures. Therefore, combining the spatial data characteristics and ecological carrying capacity of urban landscape, this paper puts forward two sustainable development paths: one is to optimize the land use layout and increase the green space and water area; The second is to promote the construction of low-carbon buildings and green infrastructure. By simulating and analyzing the spatial evolution of urban landscape under different paths, optimizing paths can effectively reduce the pressure of ecological environment and improve the sustainable development ability of cities. The data analysis shows that the optimized path 1 can increase the urban green space and water area by more than 30% and reduce the proportion of construction land by 20%, thus alleviating the pressure on the ecological environment; The second path is to reduce urban carbon emissions by about 15% by promoting green building construction. To sum up, this study provides a scientific basis for the sustainable development path of urban landscape space, and provides data support and decision-making reference for policy makers.