The Study Of X-radial Computer Tomography In The Log Non-destructive Testing

Log classifying has been a key link in the forestry producing and the lumber processing. The artificial examination has been used for a long time. This method is severely influenced by the artificial factor and the work efficiency is low, so can't be applied in the production line of the automation. This paper intends to analyze the typical defects which influences the log processing and puts forward a kind of adoption CT for examine the means of log nondestructive examination. Then applies the computer digital picture processing technique in the processing and analysis the log CT picture, and availably examines the defects of log. Carries on three-dimension reconstructions for more than 100 CT pictures of the log finally. This is significant for fixing position of the defects of the log.The log non-destructive testing technique, CT and MATLAB are had to say all first. Then relatively thorough expatiates on base of picture processing and three-dimension reconstruction technique, and carries on the summary comment to the main method of the existing log defect examination. Then important meaning of non-destructive examination applied in log procjess is clarified. These all have done the technique preparation for the design of this system. Aiming at the textual research object special, the new technique and experiment methods of picture handle and three dimension reconstruction processes are putted forward. According to the different defect in log, this paper carries on histogram balance, fast average filter, fast mediate filter, differentiation sharp, Laplace sharp operator, Sobel sharp operator and Kirsch sharp operator, etc to deal with image and distill defects. Carries on three dimension reconstructions finally. By contrast analyzing the experiment result, this text achieves the examination of the log defects and the structures successfully, and detailed argues the CT examination principle. The feasibility and rationality of CT examination is testified. This is a new way in image processing technique. At the same time, we believe it to be significant in computer pattern recognizing of defects in logs.