Research On Image Segmentation Quality Evaluation Of Wood Sheet Surface Defect Based On Gray Theory

The paper regards the application of the grey theory as the center, and focuses on two aspects, which are the grey theory’s application in digital image processing and image segmentation quality evaluation. And the paper has three main tasks, each of which is explained in detail as follows.1) Searching for the application of grey system theory in the wood defects digital image processing, which contains the grey correlation adaptive mean filtering algorithm, the images’ edge detection technology based on the correlation degree with changeable right, the fuzzy entropy threshold segmentation algorithm based on the grey correlation degree, the grey clustering image segmentation quality evaluation algorithm and the image segmentation quality evaluation algorithm based on FD grey correlation.2) Introducing and comparing some of the digital image processing technologies which is commonly used in the current, In the same way, evaluating and comparing the segmentation result of the existing wood defect image segmentation algorithm using the grey theory. In this way, we can get the two evaluation results which are obtained by the evaluation algorithm and the eye observation do match or not.3) Aiming at the characteristics of some algorithms, the paper proposes some improved schemes, and uses the grey theory evaluation algorithm to verify if the performance of the improved algorithm is superior to the traditional algorithm or not. Based on this, the paper aims for proposing improved scheme or new algorithm.After the completion of the three major tasks, the paper designs a simple image segmentation quality evaluation system and gives the GUI interface implementation using the graphical user interface development environment of Matlab2013b software.Through a large amount of experimental verification in the wood defect image segmentation and segmentation quality evaluation, we can evaluate the quality evaluation of image segmentation algorithm based on the grey theory evaluation results are consistent with the results based on the observation of the human eye or not, and if it can make a reasonable evaluation of the segmentation algorithm performance or not. And it has a significant and profound significance for selecting an appropriate image segmentation algorithm in practical application or targeted to improve existing algorithms.