Abstract: When the alternating current field measurement(ACFM) technology is used for defect determination, there are some problems such as traceability of detection data and difficulty in determining defects on site. The characteristics of ACFM inspection signals is analyzed in this thesis, and the method is proposed which deploys online data storage, inspection information display, and intelligent determination of defects method on the cloud server. The detection system is mainly composed of a detector and a cloud server. The instrument collects detection signals and transmits the detection information to the cloud server in real time. The cloud server stores the detection information and displays it on the web. At the same time, the detection information is analyzed by the defect judgment algorithm based on long short-term memory neural network (LSTM), and the results are is returned to the detector. Taking the aluminum plate specimen as the detection object, the functional test of the experimental system is carried out. The experimental results show that the developed online defect determination algorithm achieves the goals of data storage, information viewing and defect determination in the ACFM detection system.