Research On The Measurement Of Sieve Based On Optical Image

Sieve is of great significance in Industrial production and scientific research which is generally used in benchmark particle filter and powder detection in pharmaceutical industry,biotechnology and aerospace materials,etc.And how to Perform high-precision and high-speed test on test sieve is one of the most important Problems in dire need of solution for production and measurement departments.The measurement of the mesh and wire diameter size is based on the national testing standards,In view of the rectangular characteristic of the sieve and working environment.The modern digital image processing method which is efficient and objective is applied in this article to get the accurate size.This article firstly introduces the research status at home and abroad about the measurement of sieve and digital image processing methods.It also determines the overall research plan of sieve measurement system,by building good experimental environment and using industrial CMOS camera and telecentric-lens to get the clear images.In order to eliminate the impact of the uneven illumination,different algorithms are simulated to preprocessing the images and the Homomorphic filtering is adopted to improve the Image quality and remove noise.Using the gray-level histogram combining Iterative Thresholding methods to get binary images,Using morphology theory to weaken the effect of the glitch and thin hole of sieve.This paper brought up an edge detection algorithm which uses Canny operator combined wavelet transform.This algorithm gets better results than traditional algorithms between edge detection and noise suppression.In order to get the more accuracy edge positioning information.The sub-pixel edge detection algorithm is used.We use the linear least squares linear regression method combining Hough Transform to fit the discrete points of the edge in order to fit the edge of mesh and wire.Thus we can get the pixel distance between lines by pixel scanning.Finally,we use the calibration coefficients to figure out the real size of test sieve.It is proved that the implemented system based on the measurement algorithm is better than traditional methods and overcomes the traditional low-pixel edge positioning accuracy,sensitivity to noise and other shortcomings.It meets the requirements of sieve detection very well.