The invention provides an intelligent daily quality control method for LYSO crystal PET, and relates to the technical field of PET quality control. The method comprises the steps: intelligent data collection: collecting LYSO background data according to preset data collection parameter information, and storing the LYSO background data; data intelligent analysis and processing: performing statistics on information in the data to obtain a three-dimensional heat statistical graph, sending the three-dimensional heat statistical graph to a pre-trained deep learning neural network for diagnosis, obtaining an abnormal diagnosis result, and generating a DailliyQC report; and data decision and early warning: classifying the diagnosed abnormal conditions, judging whether the abnormal conditions influence the overall performance and reach the danger level or not, giving a danger alarm for the abnormal conditions reaching the danger level, and recording the abnormal conditions which do not reach the danger level and do not influence the overall performance. According to the method, an extra radioactive source is not needed, on-site operation of an operator is not needed, the data acquisition,processing, analysis and automatic identification of the DailliyQC are automatically realized, and the decisions and early warning are made.