Mainboard quality detection method, system and equipment based on image recognition and medium

The invention provides a mainboard quality detection method, system and device based on image recognition and a medium, and relates to the field of quality detection.The method comprises the steps that S1, a picture set of a computer mainboard is obtained; s2, carrying out image preprocessing on the picture set; s3, calibrating the image set after image preprocessing by using a labeling tool to obtain a sample training set and a test data set; s4, performing sample classification training based on the sample training set to obtain a model evaluation index; s5, predicting the test data set based on the model evaluation index, and selecting prediction results through an NMS non-maximum suppression algorithm to obtain an optimal prediction result as a mainboard quality detection result; the problems existing in manual quality inspection can be solved, the loss is reduced, and the quality inspection efficiency of the mainboard is improved. 本发明提出了一种基于图像识别的主板质量检测方法、系统、设备和介质，涉及质量检测领域，其方法包括：S1、获取计算机主板的图片集；S2、对图片集进行图像