Research On Measurement Method Of Flake Granules Thickness Based On Image Enhancement

There are many methods to measure granule's thickness, but the methods to the granules with inflammable and explosive properties are limited. The laser diffraction method can obtain measuring results quickly, but it is easy to produce explosion. The manual test with micrometer method can measure the thickness of granules with the slowly process, and the workers must contact granules directly. It may change the properties of granules, and do harm to the workers. Compare with the laser diffraction and micrometer method, the image processing, which becomes the development trend of measurement technology, is safer, faster and non-contact. This paper used image processing method instead of micrometer to study the micro flake energetic granules, and the measurement results achieve the industrial requirements.In this paper, the granules measuring system is built, it is composed of the image acquisition device and the software of granules measurement. The image acquisition device contains a conveying device, a granules sampling device, a camera, a light source and a chassis. The conveying device uses pneumatic explosion and controls button to transport granules repeatedly. The collection device increases the probability of falling into the range of the camera's view and solves the problem of granules bouncing and stacking by increased the screen pillars and reduced the length of the receiving station. The high quality image can be acquired through the camera, light source and the chassis with particular color and size.The measurement software includes image enhancement, camera calibration based on OpenCV, image segmentation, granues extraction and thickness measurement. The image enhancement uses the fusion method of wavelet transform and morphology to eliminate the shadow, burr and other pollution on the stage, enhance the granues' profile, improve the image contrast. The curvature driven quasi bilinear interpolation technique is introduced to interpolate the image, which is similar to the original image, and can get precise image pixel information. The camera calibration is eliminated by correcting the image, and gets the proportion coefficient of the physical size and the pixel size. Image segmentation is used to separate granues from background. The granues size is obtained by the minimum external rectangle, and the actual physical size of the granues is equal to pixel information by calibration coefficient.By analyzing the experimental data, it shows that image processing method can replace the micrometer to get the granues thickness information, and it can improve the accuracy of measurement system by improved the equipment and algorithms.