Semi-supervised equipment defect and safety monitoring image labeling method

The invention relates to a semi-supervised equipment defect and safety monitoring image labeling method. The method comprises the following steps: step 1, training to obtain an equipment defect or safety monitoring target detection model; 2, obtaining an automatic labeling result of the system; step 3, carrying out manual correction on an inaccurate labeling result; 4, taking the corrected accurate labeling information storage file and the corresponding equipment defect or safety monitoring image as input, and retraining the equipment defect or safety monitoring target detection model; 5, repeating the step 2 to the step 4, and iteratively updating the model; and step 6, embedding the target detection model which is obtained by final training and reaches the precision into an image labeling system, and completing labeling work of residual equipment defects or safety monitoring images. According to the method, a deep learning target detection model and iterative image labeling work are effectively combined, an