Instance Segmentation Method of Adherent Targets in Pig Images Based on Improved Mask R-CNN

Aiming at the segmentation problem of adherent targets in pig images, thus an instance segmentation method based on improved Mask R-CNN is proposed to separate adherent pig body. Firstly, ResNet-50 is selected as the feature extraction network of Mask R-CNN due to the single detection category. Secondly, in order to strengthen the connection between the feature extraction channels and improve the ...