Tobacco bale lining defect detection method based on machine vision technology

The invention discloses a tobacco bale lining defect detection method based on a machine vision technology, and detects the problems that a piece of aluminium-foil paper and a piece of lining paper are defected in a small tobacco bale packing process of a hard cigarette and the like. The tobacco bale lining defect detection method comprises the following steps of generating obvious difference when the aluminium-foil paper or the lining paper arranged on the inner surface of a tobacco bale is defected by adopting an industrial camera and an infrared light source with a special installation angle compared with the normal tobacco bale; and selecting multiple feature and key interest regions, and greatly reducing the difficulty of picture processing under the premise of guaranteeing that the final judging result is accurate. The tobacco bale lining defect detection method provided by the invention has the advantages that the detection accuracy is very high, the aluminium-foil paper or the lining paper defective tobacco bale is effectively avoided flowing into a market, and the application space is wide.