摘要：Nowadays, X-ray computed tomography (CT) wood nondestructive testing technology has been applied to detection of internal defects in log for the purpose of obtaining the optimal wood cutting plan. Multifractal spectrum and wavelet transform are usually used for analyzing, modeling, and extracting different complex features of signals and images. A novel CT image edge detection method which using multifractal spectrum theory combined with wavelet transform is applied in this paper. The new method can be divided into the following main steps: (1) Calculating the wavelet module values of wood defect image. (2) Combining wavelet transform module values with multifractal theory. (3) Calculating the multifractal spectrum from the wavelet transform. (4) Selecting the appropriate threshold to wood defects detection. A large number of experimental results show that the new method to recognize the wood defects is effective.
