Intelligent detection method for missing cigarette sticks of a box sealing machine based on similarity analysis and feature incremental learning

Abstract

The invention discloses an intelligent detection method for missing cigarette sticks of a box sealing machine based on similarity analysis and feature incremental learning. In the off-line modeling phase, the gray level transformation and image segmentation are carried out on the front and back images of cigarette sticks collected from the cigarette packets, and 16 feature similarity indexes of the modeling images are calculated by using the similarity analysis method of horizontal and vertical partition, and the thresholds of 16 features of the modeling images are determined by using the kernel density estimation method. In the online recognition stage, the similarity analysis method of horizontal and vertical blocks is used to calculate the front and back images of cigarette bars in thecurrent cigarette packets in real time, and 16 feature similarity indices of the current images are obtained. By comparing with the current threshold, whether the front and back images are missing ornot is recognized. The invention realizes the effective characterization and quantification of the cigarette bar picture characteristic information in the cigarette pack, can accurately judge the occurrence of the cigarette pack missing bar quality defect, and provides the information basis for the proof of the missing bar through the picture marking.

Description

technical field [0001] The invention relates to the field of detection of missing strips in cigarette packs of case sealing machines, in particular to an intelligent detection method for missing strips of cigarette packs of case sealing machines based on similarity analysis and feature incremental learning, which is suitable for cigarette case sealing machine equipment. Background technique [0002] Cigarette packing is the last process of cigarette production. During the packing process, the quality of cigarette packs may be missing. Most cigarette companies have strengthened process monitoring and management. Through the transformation and upgrading of the sealing machine, electronic weighing heavy-duty strip detection methods or machine vision missing strip detection methods are used to detect the quality problems of cigarette pack missing strips that may occur during the packing process. . With the promotion of cigarette carton recycling projects in the tobacco industry...

AI Technical Summary

Problems solved by technology

With the application of new detection methods and systems for missing rods, the quality problems of missing rods have been reduced, but they have not been completely cured. Almost every year, there will be several or even dozens of complaints about missing rods in cigarette factories, which will negatively affect the brand image of the company.

Images