基于多源卫星遥感影像的广西中南部地区甘蔗识别及产量预测

罗维, 李修华, 覃火娟, 张木清, 王泽平, 蒋柱辉

Identification and yield prediction of sugarcane in the south-central part of Guangxi Zhuang Autonomous Region, China based on multi-source satellite-based remote sensing images

LUO Wei, LI Xiuhua, QIN Huojuan, ZHANG Muqing, WANG Zeping, JIANG Zhuhui

表6 各网络最佳模型的评价指标

Tab.6 Evaluating indicators of the best models

模型 Kappa 总体精 度/% 查准 率/% 查全 率/% 单张推理 时间/ms BiseNetV2 0.798 6 90.43 94.62 89.87 48 ECA-BiseNetV2 0.806 9 91.54 95.57 90.78 50 U-Net 0.681 5 83.67 94.61 77.13 101 SegNet 0.642 7 81.95 95.69 73.84 95 DeepLabV3+ 0.801 7 91.09 94.90 90.36 105 Fast-SCNN 0.675 2 84.19 92.32 81.52 39 BiseNetV1 0.744 8 87.68 94.27 85.51 55