Magnetic particle flaw detection is one of the most sensitive detection methods for surface and near-surface defects of ferromagnetic materials.Due to its high requirements on the experience and technology of inspection personnel,it is easy to cause missed detection and wrong detection in daily inspec-tion.Intelligent defect recognition method can greatly reduce the probability of missing detection and wrong detection.The automatic algorithm module of magnetic mark defect image recognition based on BP neural network can train the feature of magnetic mark image from multiple angles and different features to meet the requirements of magnetic mark image detection in different environments.
关键词
磁粉检测/磁痕显示/BP神经网络/图像识别
Key words
Magnetic particle flaw detection/Magnetic mark display/BP neural network/Image recogni-tion
维普
万方数据