Deep learning-based drilling process automatic identification supervision method and system

The invention discloses a drilling process automatic identification and supervision method and system based on deep learning, and the method comprises the steps: obtaining a monitoring video stream of the drilling process of a target polluted site through machine vision, and extracting key frame image data for preprocessing; obtaining a region of interest, identifying model information of the drilling equipment by using the region of interest, extracting a drilling action data set of the drilling equipment according to the model information, and combining drilling action data with the single drilling depth; constructing a drilling identification supervision model, importing a region-of-interest sequence, identifying a single drilling action, reading a corresponding single drilling depth, and outputting a final drilling depth in a drilling process; and comparing the drilling depth with a preset drilling depth to generate drilling depth abnormal information. According to the method, the drilling process is effe