| Automatic visual inspection system has the advantages of non-contact, high speed, high precision, and high on-site anti-interference ability.It can effectively overcome the shortcomings of manual inspection and improve the degree of automation and production efficiency.With China’s economic development, changes in the economic structure and improvement of labor costs and product structure, the requirements for automatic visual inspection system has become very urgent.With the development of computer technology, the application of image processing in automatic visual inspection is more and more widely. Traditional image processing method based on the airspace and frequency domain has various shortcomings. To improve the effectiveness and speed of operation of the image processing algorithms is especially meaningful.The subject of the study is the mixture of cut tobacco and terrier removed from the rolling process in cigarette factories.This article uses digital image processing technology, researches recognition algorithm of cut tobacco and terrier, and designs the overall system program.The main work is:1、Design a detection system of cut tobacco and terrier.It mainly consists of light, CCD camera,image acquisition card,and image processing module.2、Discuss pre-processing algorithm of the captured image, including image filtering, enhancement algorithm, and then write software to test the effectiveness of the algorithm on a PC.3、An improved Otsu method based on the traditional image segmentation algorithm is used, and it verifies that the algorithm can achieve better segmentation results by experiments.4、Morphology thinning algorithm based on two sets of structural elements template is used on the basis of the definition of contour and skeleton. The image of cut tobacco and terrier is thinned by this algorithm,obtaining satisfactory results. 5、In MATLAB7.0environment, image processing algorithms is studied, and the cut tobacco and terrier detection software is developed. |
PDF Full Text Request