Research on automatic recognition of defects in radiographic images of liquid chlorine tankers based on deep learning
The Radiographic image of liquid chlorine tanker may contain many different types of defects,and the manifestation of each defect in the image may be quite different,which leads to the phenomenon of low number of defects identified.Aiming at the above phenomenon,the research on automatic recognition of defects in Radiographic images of liquid chlorine tank cars based on deep learning is proposed.The Radiographic image of liquid chlorine tanker was preprocessed to eliminate noise and enhance image characteristics.The convolution layer,pool layer,full connection layer and classifier in the deep learning model were designed,and the defect identification model of liquid chlorine tanker Radiographic image was established,and the model was trained.The output of the model was calculated by forward propagation algorithm,and the error was traced back layer by layer by backward propagation algorithm,and the weight parameters of the model were optimized.The automatic defect identification of liquid chlorine tanker Radiographic image was realized by adopting adaptive learning strategy and iterative training.The experimental results show that the research method can effectively identify various types of defects in the Radiographic image of liquid chlorine tanker,and the number of defects is high.
deep learningliquid chlorine tank carradiographic imageimage defectautomatic defect identification
万方数据
维普
