Automatic driving sanitation vehicle target detection model

An automatic driving sanitation vehicle target detection model belongs to the technical field of target detection, and comprises a direction sensitive attention sub-network embedded in an original target detection model, and the original target detection model processes image data through an original convolutional layer thereof to obtain a plurality of original feature maps; the direction sensitive attention sub-network carries out direction feature extraction on the plurality of original feature maps through a direction convolutional layer of the direction sensitive attention sub-network to obtain weight coefficients representing a plurality of set directions in each original feature map; and the original target detection model performs target detection according to the plurality of original feature maps and the corresponding weight parameters to obtain a target detection result. According to the target detection method and device, the direction sensitive attention sub-network can be used for replacing a par