In order to monitor the status of digital picture book image sensors and perform fault diagnosis,a cloud platform based remote fault monitoring system for digital picture book image sensors was studied and designed.The system utilizes rough set theory for data reduction,and uses a Bayesian network model to calculate the maximum probability of fault causes and infer the cause of sensor faults.The experimental results show that the research model took 1.83 minutes and 2.21 minutes to process 400 fault data on IBM Watson Studio and SAS Enterprise Miner software,with stable operating values of about 91％and 94％,and fault diagnosis accuracy of 92.3％and 93.2％,respectively.This indicates that the model can effectively monitor sensor status and analyze the cause of faults,and provide a reliable theoretical and methodological basis for sensor fault diagnosis.
维普
万方数据