Application of Multi-Scale Fractal Feature in Defects Detection of Log X-Ray Image

The aim of this study is to extract the internal defects information from a log x-ray image and accurately identify the types of the defects in log by digital image processing technology and fractal theory. A method in log x-ray image defects detection based on multi-scale fractal feature (DMF) was applied in this paper. The DMF values of different regions in a log image are normally different. According to the values of DMF, the internal defects in log can be detected and classified. This method has been applied to the log image with crack and knots, the experimental result show that the DMF values of the normal regions are between 0.020 and 0.060; the DMF values of the crack regions are between 0.300 and 0.600; and the DMF values of the knot regions are between 0.100 and 0.200. This study provides a new tool for analyzing and processing log x-ray image.