Edge calculation motor oil stain identification method based on deep learning

The invention discloses an edge calculation motor oil stain identification method based on deep learning. The method comprises the following steps: S1, making a sample data set; s2, completing training and clipping of a deep learning model in Amazon SageMaker, namely, introducing Focal loss into a confidence loss function of a YOLOv3 algorithm, and training and clipping the deep learning model by using the loss function; s3, remotely deploying the deep learning model to the edge device which builds the AWS IoT Greengrass environment; and S4, importing an industrial machine picture shot in real time into the edge equipment, automatically identifying whether oil leakage occurs or not through a deep learning model, and outputting a result. According to the oil stain identification method, the purposes of training, clipping and reasoning prediction separation of a deep learning model are achieved, the deep learning model is trained and clipped at the cloud, and the clipped deep learning model adapts to resources