Equipment anomaly detection system and method and electronic equipment

The invention discloses an equipment anomaly detection system and method and electronic equipment. The system comprises an attribute set identification module used for obtaining attribute information according to input target image data; the data attribute graph construction module is used for obtaining an attribute graph of the target image data according to the target image data and the attribute information thereof; the attribute auxiliary feature refining module is used for performing feature aggregation on the attribute graph to obtain deep features of the target image data and further obtain predicted target graph line data with a pseudo tag; and the detection module is used for training according to the predicted target graph line data to obtain a trained detection module. Through deep feature extraction of target image data, high-quality predicted target graph line data with pseudo labels are obtained, data labeling is completed efficiently and quickly, and a large amount of training sample data can b