The invention discloses a method for detecting the temperature abnormal region of a crystallizer copper plate based on a cellular automaton, belongs to the field of ferrous metallurgy and mainly solves the problems of difficult capture of heat transfer abnormal and low search efficiency of abnormal region features in a crystallizer under severe working conditions. According to the technical scheme, the method comprises the following steps: (1) the temperature of the crystallizer copper plate is visualized; (2) initializing the cellular automaton model; (3) communicating and merging the cellular markers of the region; (4) abnormal cellular region boundary detection. According to the invention, the cellular automata method is combined with the detection of the temperature abnormal region ofthe crystallizer, the regional connectivity is judged by utilizing the temperature state of the current cell and the near-domain cell, and the region boundary is obtained through the starting cell andthe terminating cell of the same reference number region communication segment, hot spots, cold spot areas and boundaries of the mold bonding, cracks and other abnormalities are rapidly distinguishedand extracted, and an accurate and reliable image detection method for prediction of steel leakage, longitudinal crack of the casting blank, angular cracks and the like is provided.