摘要
论文利用深度学习卷积神经网络(CNN)模型,在提取建设用地时空特征的基础上,对这些样本数据进行训练和验证,实现了对建设用地时空演变特征的精准识别和分析。研究结果表明:随着城市化进程的不断推进,南沙新区的建设用地呈现出明显的时空演变特征,表现为耕地和林草覆盖面积逐渐减少,而建设用地面积显著增加,城市用地扩张速度加快、建设用地类型结构不断优化等方面的变化,这些结论对于深入理解南沙新区的城市化发展过程,指导城市规划和土地利用具有重要意义。
This study employs the deep learning Convolutional Neural Network(CNN)models to extract spatial-temporal features of construction land and to train and validate the sample data,achieving precise identification and analysis of the spatial-temporal evolution characteristics of construction land.The research results demonstrate that with continuous advancement of urbanization,the construction land in Nansha New District exhibits distinct spatial-temporal evolution characteristics.These are characterized by a gradual decrease in the area of arable land and vegetation cover,a significant increase in the area of construction land,an acceleration in the speed of urban land expansion,and continuous optimization of the structure of construction land types.These findings are of significant importance for gaining a deeper understanding of the urbanization process in Nansha New District and for guiding urban planning and land use.
作者
孟凡纪
尹小杰
MENG Fanji;YIN Xiaojie(Dongguan Research Center of Geographic Information and Planning,Dongguan 523000,China)
出处
《江西测绘》
2025年第1期18-21,34,共5页
JIANGXI CEHUI
关键词
长时序卫星遥感
卷积神经网络模型
时空演变特征
Long-term Satellite Remote Sensing
Convolutional Neural Network Model
Spatial-temporal Evolution Characteristics
