A Gender Detection Method for Crayfish Based on YOLO v5
In response to the low efficiency and high cost of traditional manual discrimination of the gender of crayfish(Procambarus clarkii),this study proposed a gender detection model for crayfish based on YOLO v5,achieving automatic discrimination of the gender of crayfish.Crayfish images were captured using a self-designed device,and data annotation was performed using two methods,i.e.,detection based on the petasma and detection based on regional features,using the Labelme tool.A binary classification model was trained using Resnet-18 as the feature extraction network under the Pytorch framework,and two models,one for petasma detection and the other for regional feature detection,were trained based on YOLO v5.The results showed that the model based on regional feature detection had high detection performance and accuracy,efficiently and cost-effectively extracting the gender characteristics of crayfish,which was of great significance for the breeding improvement of crayfish.
NETL
NSTL
万方数据
维普
