To avoid colliding with trees during its operation,a lawn mower robot must detect the *** tree detection methods suffer from low detection accuracy(missed detection)and the lack of a lightweight *** this study,a datas...
详细信息
To avoid colliding with trees during its operation,a lawn mower robot must detect the *** tree detection methods suffer from low detection accuracy(missed detection)and the lack of a lightweight *** this study,a dataset of trees was constructed on the basis of a real lawn *** to the theory of channel incremental depthwise convolution and residual suppression,the embedded-A module is proposed,which expands the depth of the feature map twice to form a residual structure to improve the lightweight degree of the *** to residual fusion theory,the embedded-B module is proposed,which improves the accuracy of feature-map downsampling by depthwise convolution and pooling *** embeddedyolo object detection network is formed by stacking the embedded modules and the fusion of feature maps of different *** results on the testing set show that the embeddedyolo tree detection algorithm has 84.17%and 69.91%average precision values respectively for trunk and spherical tree,and 77.04% mean average precision *** number of convolution parameters is 1.78×10^(6),and the calculation amount is 3.85 billion float operations per *** size of weight file is 7.11MB,and the detection speed can reach 179 frame/*** study provides a theoretical basis for the lightweight application of the object detection algorithm based on deep learning for lawn mower robots.
暂无评论