基于Psi-Net深度学习网络的高空间分辨率遥感影像地块尺度的耕地提取  

High spatial resolution remote sensing image plot scale farmland extraction based on Psi-Net deep learning network

作　　者：马海荣 沈祥成 罗治情 陈娉婷 郑明雪 官波 MA Hai-rong;SHEN Xiang-cheng;LUO Zhi-qing;CHEN Ping-ting;ZHENG Ming-xue;GUAN Bo(Institute of Agricultural Economics and Technology/Hubei Agricultural Science and Technology Innovation Center Agricultural Economic and Technological Research Sub-Center/Hubei Rural Revitalization Research Institute,Hubei Academy of Agricultural Sciences,Wuhan 430064,China)

机构地区：[1]湖北省农业科学院农业经济技术研究所/湖北省农业科技创新中心农业经济技术研究分中心/湖北省乡村振兴研究院,武汉430064

出　　处：《湖北农业科学》2024年第11期197-202,共6页Hubei Agricultural Sciences

基　　金：湖北省自然科学基金一般面上项目(2022CFC047);国家自然科学基金青年科学基金项目(NSFC42201397)。

摘　　要：将语义分割和边缘检测深度学习网络结合,构建Psi-Net深度学习网络。结果表明,Psi-Net深度学习网络能有效识别耕地,绝大部分耕地被有效提取出来,正确率(Accuracy)为96.3%,生产精度(PA)为98.1%,用户精度(UA)为97.1%。Psi-Net深度学习网络在耕地地块边界识别时有了耕地范围的限定,减少对非耕地地块边界的识别,完备性为74.3%,正确性为80.2%,质量为62.8%。Psi-Net深度学习网络可以有效识别面状耕地范围,并且在耕地范围的限制下,提取的地块尺度耕地边界均落在耕地范围内,不会对耕地外的地块边界进行识别,有效减少了似地块边界提取噪声的影响。Combining semantic segmentation and edge detection deep learning networks,a Psi-Net deep learning network was constructed.The results showed that the Psi-Net deep learning network could effectively identify cultivated land,and the vast majority of cultivated land was effectively extracted with an Accuracy of 96.3%,a production accuracy(PA)of 98.1%,and a user accuracy(UA)of 97.1%.The Psi-Net deep learning network had limited the scope of cultivated land for boundary recognition,reducing the recognition of non-cultivated land boundaries.The completeness was 74.3%,the correctness was 80.2%,and the quality was 62.8%.The Psi-Net deep learning network could effectively identify the range of surface cultivated land,and under the limitation of cultivated land range,the extracted plot scale cultivated land boundaries all fell within the cultivated land range,without recognizing the plot boundaries outside the cultivated land,effectively reducing the impact of noise in extracting plot boundaries.

关 键 词：Psi-Net深度学习网络 高空间分辨率 耕地地块 遥感影像 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]