Study on the Key Image Processing Technology in the Inspection of Packing Quality for Small-Pack Cigarettes

Image processing is a key factor in the inspection of packing quality by machine vision. Targeting the inspection of packing quality for small-pack cigarettes in tobacco enterprises, this paper proposes a series of key image processing techniques: a fast algorithm for small-pack edge collecting based on fast Hough transformation, an online differentiating algorithm based on GLCM parameters, and packing-flaw detection based on SVMs, etc. The key techniques are applicable to the inspection of external product-packing quality, and have engineering referential value for constructing product inspection systems based on machine vision.