改进Deeplabv3+的高分辨率遥感影像道路提取模型

赵凌虎, 袁希平, 甘淑, 胡琳, 丘鸣语

An information extraction model of roads from high-resolution remote sensing images based on improved Deeplabv3+

ZHAO Linghu, YUAN Xiping, GAN Shu, HU Lin, QIU Mingyu

表3 不同模型对比结果

Tab.3 Comparing the results for the different models

模型 OA/% P/% R/% F1/% 参数 量/M 平均 帧数 PSPNet 96.59 63.59 77.16 69.72 2.38 129.56 U-Net 98.26 82.46 82.33 83.39 24.89 49.71 Deeplabv3+ 98.45 87.00 81.92 84.38 54.71 30.77 改进Deeplabv3+ 98.71 87.49 87.06 87.27 5.81 120.05