Segmentation Technology Of Wood Defects Based On Mathematical Morphology

With the development of computer technology, some scholars at home and abroad using computer vision technology and image processing techniques for wood surface defect image processing, trying to achieve timber defects of fast and stable detection methods to replace the traditional manual testing, computer-controlled wood defect automatically. However, the plate surface defects in the type, size, shape and color of the complexity and diversity, and the wood texture, using various methods of image segmentation results are not satisfied, often exist in the segmented image some of the texture primitive edge of the particles.Mathematical morphology applied to wood defect image segmentation technology has become one of the main content of research in recent years, wood nondestructive testing process.The papers mainly focus on the morphological segmentation techniques of edge detection algorithm for processing and analysis of three common wood defects insect eye, knot, slipknot, in order to more in-depth study. The paper first describes the classification of wood defects, the application of mathematical morphology in image processing technology, wood nondestructive testing technology development in the world; then study the basic principles of image segmentation; and then from the binary image, gray scale images and color the image is followed by in-depth, specific research on timber defects.On one hand, the morphological segmentation algorithm based on edge in the range of grayscale images, Part in the edge-based image segmentation, morphological advantage in image edge detection, a comprehensive, multi-scale morphological structuring elements are defined by a weighted combination of morphological operations, a full range of multi-structure, multi-scale edge detection method. Defect image will be compared with directly by the morphological operation of binary defect edge and traditional edge detection algorithm is further evidence of the noise resistance and practicality of the method.On the other hand, in-depth study of a binary image, gray-scale image a variety of morphological analysis algorithm based on mathematical morphology extension to color images. Edge detection in color images, the extraction of the target defect image not only with the detection algorithm, the classification of the different color space also affects the test results. RGB color space using only the brightness information and less use of color information, and the space three components are highly correlated, human vision prefers to use hue, saturation and brightness to describe the color object is inconsistent, not suitable for edge detection task. RGB image is converted into the three components of the HSV images were of different color space multi-structure, multi-scale morphological filtering and comprehensive form of edge detection. Ultimately, Wood defect detection in binary images, grayscale images and color images, experimental results are analyzed and discussed, the results show that the extraction of timber defects in the color space methods to take full advantage of the color information of the image, the more complete retention of the contours of the original color image, than under the two space edge detection method, the results obtained in the inhibition of the noise, color variation, has greatly improved.