Study On Real-Time Non-Destructive Testing For Woody Material Based On Mathematical Morphology

China is a large population country with relatively scarce resources and weak environment carrying capacity. Now, the contradiction between the fast loss of China's forest resources recoverable accumulation amount and the urgent need for rapid development of the national economy is still quite big. Protection of existing forest resources, scientific and full utilization of wood is an important way to resolve this contradiction. Therefore, the action of seeking a more scientific method for nondestructive testing woody material and obtaining the right shape, size and defects accurately before sawing can improve wood utilization and economic efficiency, reduce production costs and conserve forest resources. The action has important scientific value and practical significance.The purpose of the study was to build a real-time nondestructive device for woody material detection, and to apply optimized digital image processing techniques to woody material defect detection for the high detection accuracy. Detailed analysis, summary and comparison in the nondestructive testing of wood defects in the current situation and development trend were given by domestic and international status of non-destructive testing technology, and the advantages and limitations of a variety of detection methods were described. Under the existing experimental conditions the X-ray detection was selected as a nondestructive testing means of woody material defects. Digital image processing was used to improve wood defect recognition accuracy. And improved mathematical morphology operator was applied to woody material real-time nondestructive testing device.On the basis of original wood X-ray detection system, a real-time imaging nondestructive testing hardware system was designed for woody material detection, and the stability and viability of device was tested in the experiment. The testing hardware system achieved continuous, real-time, on-line detection, full range of multi-angle detection and improved the image quality and measurement accuracy for woody material detection by using high-frequency high-voltage multi-focal X-ray, image intensifier and designing the mobile detection platform, rotating platform and mechanical arm. Three-window console in the device realized centralized control for circuits and mechanical devices, which saved manpower and operation hours. In order to reduce ray radiation hazards, the experimental environment was optimized design and transformation enhancing the protective effect.Advanced image acquisition card MATROX was use as the computer hardware equipment for collecting woody material X-ray images. Image processing software work platform for woody material non-destructive testing in real-time imaging system was developed and adapted to ordinary PC in Windows environment. The software could realize the dynamic acquisition and static processing of woody material X-ray images, which could provided a new non-destructive detection method in the quality identification of wood board and defects detection in the logs.In order to increase the accuracy of image recognition and obtain a clearer X-ray image of woody material, histogram equalization and histogram specification processing were used to enhance image contrast. Average filtering and median filtering method were used to suppress the noise in the woody material images X-ray image. Differential operator, Laplacian-Gaussian operator and Canny operator were used for the edge detection of woody material X-ray images. The preprocessing of woody material images provided the basis in image processing by mathematical morphology approach.Based on the basic theory and algorithm of mathematical morphology, self-adaptive multi-scale weight morphological operator was advanced and used in the defect identification of woody material X-ray image. As can be seen in the experimental analysis, compared with the traditional edge detection algorithm, self-adaptive multi-scale weight morphological operator had the characteristics of a high degree of detection and identification accuracy in woody material edge detection.Woody material non-destructive testing real-time imaging hardware and woody material real-time imaging image processing software system were built in the study. And self-adaptive multi-scale weight morphological operator was used to edge detection of woody material X-ray image. The achievements in the study realized the real-time online detection of woody material X-ray image, and improved the detection accuracy of defects in the image. They have a wide range of applications in the quality identification of wood board and defects detection in the logs.