The invention provides a tooth mark detection method and device, electronic equipment and a readable storage medium, relating to the technical field of image processing. The tooth mark detection method is based on the deep learning neural network, and the deep learning neural network is trained through the sample tongue image, so that the deep learning neural network has the capability of identifying the tooth mark region in the tongue image, and the tooth mark region in the to-be-identified image can be identified. The trained deep learning neural network is adopted as a main body of tooth mark recognition. Manual participation is not needed in the recognition process. No empirical value is needed. Tooth mark area detection can be automatically completed, and detection accuracy is higherthan that of a decision classifier, a curve fitting method and a concave point detection method.