基于上下文增强与多目标语义感知的电力缺陷检测算法

胡欣,阎希玥,常娅姝,程鸿亮,肖剑,罗诗伟,马亮

Context-enhanced multi-target semantic perception algorithm for power defect detection

Xin HU,Xiyue YAN,Yashu CHANG,Hongliang CHENG,Jian XIAO,Shiwei LUO,Liang MA

表 3 高低频卷积模块通道占比取值实验

Tab.3 Channel-ratio experiment of high-low frequency convolution module

数据集 $ \alpha $ P R mAP50 mAP50-95 VOC 0 0.671 0.568 0.596 0.451 0.125 0.703 0.570 0.611 0.466 0.250 0.701 0.578 0.616 0.466 0.500 0.697 0.573 0.615 0.464 0.750 0.685 0.569 0.601 0.456 Electric_Data 0 0.816 0.660 0.691 0.393 0.125 0.814 0.693 0.701 0.389 0.250 0.856 0.695 0.704 0.395 0.500 0.811 0.666 0.706 0.395 0.750 0.827 0.666 0.691 0.386