引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
过刊浏览    高级检索
本文已被:浏览 39次   下载 0  
分享到: 微信 更多
基于U-net的甘蔗深度学习分类方法初探
董秀春, 蒋怡, 王思, 李宗南, 王昕
四川省农业科学院遥感应用研究所
摘要:
【目的】为实现使用高分辨率遥感影像快速提取作物种植空间信息,探索基于深度学习的甘蔗分类方法。【方法】以云南省陇川县甘蔗种植区为研究区,收集0.5m的Google earth开放影像,建立训练区,训练U-net深度学习分类模型参数;使用U-net深度学习分类器提取甘蔗种植空间信息,通过地面样方数据初步验证该方法的精度。【结果】(1)基于深度学习方法的甘蔗分类总体精度和Kappa系数分别为92.76%和0.8480,面积总精度为94.41%;平坝区、丘陵区分类精度存在差异,总精度和Kappa系数分别为97.10%、0.9221和88.42%、0.7673;(2)受部分地物RGB影像特征与甘蔗相似的影响,分类结果存在部分的错分现象。【结论】深度学习方法在提高高分辨率影像的甘蔗分类精度和效率具有极大潜力,更准确的分类精度还有待于相关研究和验证。
关键词:  遥感;深度学习;作物分类提取;甘蔗
DOI:
分类号:
基金项目:四川省农业科学院前沿学科研究基金“基于遥感大数据和深度学习的作物种植信息提取”(2019QYXK036);四川省科技厅应用基础研究项目“基于空间大数据的乡村地区土地利用变化研究”(2019YJ0608);四川省财政创新能力提升工程专项资金项目“成都市天府新区智慧农业研究”(2016GXTZ-011)
Preliminary study on deep learning classification of sugarcanebased on the U-net model
Dong Xiuchun, Jiang Yi, Wang Si, Li Zongnan, Wang Xin
Institute of Remote Sensing Application,Sichuan Academy of Agricultural Sciences
Abstract:
[Purpose]To extract the desired information from remote sensing big data requires efficient data processing capabilities. In order to realize the extraction of crop planting spatial information by using high-resolution remote sensing images rapidly, a method of sugarcane classification based on deep learning was preliminary studied. [Method] In this study, part of the Sugarcane planting area in Longchuan county which located in Dehong prefecture was selected as the study area. Subsequently, the classification of sugarcane was performed in ENVI5.5 through the deep learning model which is based on the U-net architecture. The specific operations are as follows. First, the Google Earth image on November 28, 2018 was collected and resampled into an image with 0.5m spatial resolution and 8km×8km width. Second, the region of interest(ROI)about Sugarcane were selected and the ground sample verification data was produced by manual interpretation. Training work about U-net deep learning classification model parameters were completed in envi5.5. Finally, Sugarcane classification and spatial information extraction were completed by using the trained deep learning classifier, and the accuracy of the method was verified preliminary.[Result]This study revealed the following conclusions. First, the overall accuracy and Kappa coefficient of Sugarcane classification based on deep learning method were 92.76% and 0.8480, respectively, and the area accuracy was 94.41%. Second, there were differences in classification accuracy between different regions. The total accuracy and Kappa coefficient of the flat dam area were 97.10% and 0.9221, respectively, much higher than the hilly area which has the results of 88.42% and 0.7673. In addition, there are too many pixels that were misclassified and miss classified in the hilly area, especially the miss classified pixels were much more than flat dam. Third, partial misclassification were existed in the classification results by using Google Earth images, because it is difficult to distinguish in RGB image for some objects features are similar to Sugarcane.[Conclusion]This study showed that the deep learning method has great potential in improving the accuracy and efficiency of sugarcane classification using high-resolution images, and more accurate classification accuracy remains to be further studied.
Key words:  Remote sensing;deep leaning;crop classification;Sugarcane