Research On The Problem Of Ray Application In The Material Straining Experiment Based On Video Image

In recent years, the high speed development of computer technology has deepened the contact of multidisciplinary such as optics,optoelectronics,image processing,artificial intelligence and so on. It has also promoted the development of technique which concerns with vision. So it has been accepted by people. The measuring technique based on video image realizes the function of human vision—the perception,identification and comprehension to the objective world. What's more important, it can also finish the work which human vision can not do. It has the characteristics of the altitudinal brainpower and universal flexibility and it has permeated to aspect of human life and socioeconomic development. For these reasons, this technique has received high recognition from the allied nations and various vocations.In the non-contact measurement which based on video image, illumination is an important factor that impact input of the system, because it influence the quality of the input data and at least 30% application effect directly. For each specific example of application, it should select the relevant illuminating equipment in order to reach the purpose of optimum efficiency. But how to make the luminous energy remain firm in a definite degree is a problem to be solved in the utility course; On the other hand, the change of the light environment is also one of the major factors that affect the measurement accuracy. The ambient light will change the overall luminous energy irradiating on object, and so it will brought out the deferent image data with noise.In allusion to the material strain measurement based on video image, this paper studies the relationship between image quality and lighting in the imagemeasurement. The purpose is to find a reasonable,practical using method of light, solve the lighting problems in the system and improve the image quality. So it lays the foundation for the further research and the development work.At first this paper summarizes the concept and the application of the non-contact measurement technology based on video images, and explains the methods of light treatment existed. At present, the solution about the light has been divided into two angles, pre- and recognition algorithms. Since the capability of the recognition algorithm depends on the training image illumination and the practicality is not strong, how to solve the lighting problems is about statistical methods. Pretreatment methods mainly include the method based on image processing technology and the mode-based method. This paper outlines several typical algorithms and analyzes the characteristics of these methods briefly.Second, about the subject of strain measurement system, this paper introduces the working principle, the software and hardware design of the system briefly. At the same time it summarizes several key technologies—edge detection, illumination technology and light treatment. This system gets the right image contrast mainly by using LED lighting system and digital CCD camera and adjusting the distance between the camera and aperture to the changes. In the experiment, the digital image is got by CCD camera and the data is transmitted to the computer by image card. Then we process the digital image data by the related software. Finally, it can immediately and accurately export the displacement variation of the mark on the specimen. During the measurement, it is necessary to identify the edge of the specimen image from the camera accurately. On this basis, we get the change of its geometry dimension. So it is obvious that the accurate identification and pick-up has a great impact to the measurement result. In order to improve the measurement accuracy of the system, we use the sub-pixel positioning method with the first-order differential expectations. In addition, to obtain high-quality images, we need to consider the illumination technology and the correction of uneven light.Then the paper starts with the light technical. After analyzing the theory of theoptical elements and light source evaluation, we design the LED light according to the needs of the system and analyze the optical environment of the system and the distribution of the lame-house. For this measurement system, it mainly deals with the real-time image of the specimen, so the lighting source should show the feature of objects prominently, and there should be a clear distinction between the part for detection in need and those least of objects to increase the contrast. Meanwhile, the overall brightness should also be ensured sufficiently, and the features should be outstanding as soon as possible. In the light source we design, LED are arranged in a matrix form on the printed circuit board close and the optical glasses are below the printed circuit board, so it makes the ray be scattering nearly. Because the LED are put close, the light is close to the uniform optical scattering. When all the LED are brightened, it is vertical irradiation uniformity. In the next light distribution of the system, we analyze the illumination uniformity of the light source in detail. The result shows that the illumination uniformity reaches 0.949, and it is able to meet the demands of the vast majority of applications.In addition, the material property of metal may lead to uneven distribution of the illumination. In practical application, although we give sufficient attention in the design of light source, there is also environmental-impact. It will make some rather strong and some dark in the image because of uneven illumination or some other reasons. Under these circumstances, the edge detection of image and pattern recognition work can not be achieved to a satisfactory result. It is likely to get error, and lead to the reduction of measurement accuracy ultimately. Therefore, in order to reduce the influence of the intensity of illumination, each pixel in the image should be corrected when processed. Finally, in order to correct the image pixel values, we use the mutual superposition method of nonlinear function in allusion to direct lighting in the paper. So it can weaken the impact of light and improve image quality. The experimental result shows that the method has a better correction to the reflecting problem on the metal specimens and it can reduce the impact of light to some extent, and improve image quality.The research and analysis shows that now it is very necessary to study the useof the light in the non-contact measurement system based on video image. It has a direct impact to improve the image quality and the measurement precision of the system.