Single Out Of Pc Software In The System Design And Implementation

Loose-end cigarette is a widespread quality problem in domestic cigarette manufacturing companies. The accurate detection and rejection of loose-end cigarettes not only improve the quality of cigarettes, avoid loose-end cigarettes flowing into market and affecting the brand image of the cigarette manufacturing, but also can reduce the consumption of paper, filters, aluminum foil and other materials and the cost on manual inspection, so the cigarette production will be increased efficiently with good economic and social benefits. However, the mechanical contact and photoelectric detection methods that are already applied can be easily influenced by environment, and thus the detection results are unsatisfactory. How to detect and eliminate loose-end cigarettes accurately and quickly is an urgent problem.In this paper, a method based on machine vision is presented to detect loose-end cigarettes, and the design and implementation of the PC software, as a part of the detection system, is emphasized. The overall design of the PC software in accordance with the modular design concept is completed, and the key problems in its implementation are solved. The PC software communicates with the IO control panel by RS-232serial ports, and checksum is joined to ensure the reliability of the communication. The man-machine interface is organized by functional modules. Digital image processing algorithms, including Gaussian filtering, adaptive thresholding segmentation and circle detection based on random Hough transform, are used. A new algorithm for loose-end cigarettes detection is proposed based on the analysis of image data, which satisfies the demand of this system with an effective rejection rate of83.3%d and a mistaken rejection rate less than10%. Tested on experimental platform, the PC software runs steadilv.