Cigarette quality detection method based on two-stage algorithm

Abstract

The invention discloses a cigarette quality detection method based on a two-stage algorithm. The cigarette quality detection method comprises the steps of: acquiring an original image, carrying out interested region identification on the original image, extracting target images, and then carrying out graying and binarization processing; extracting largest target contours of cigarettes in the binarized target images, calculating contour areas according to the largest target contours, comparing the contour areas with an area threshold and determining unqualified cigarettes so as to implement a first stage of detection on cigarette quality; and for residual cigarettes which are not removed at the first time, carrying out matching on the cigarettes in the target images and a cigarette standardtemplate by adopting a Match Template function, comparing matching values to a matching threshold, determining unqualified cigarettes so as to implement a second stage of detection on cigarette quality, and removing the unqualified cigarettes. Cigarettes which are still not removed after the two-stage detection algorithm is carried out are determined as qualified cigarettes for preservation. By analysis and determination on the to-be-detected cigarettes and removing the unqualified cigarettes, effects of reducing production cost and improving quality of cigarette products are achieved.

Description

technical field [0001] The invention belongs to the field of quality detection, and in particular relates to a method for detecting the quality of cigarettes based on a secondary algorithm. Background technique [0002] With the continuous improvement of my country's cigarette production technology, the automation and intelligence of rolling and packaging equipment has become the future development trend of the tobacco industry. ZJ116 ultra-high-speed crimping unit has the characteristics of fast production speed and high degree of automation, and is widely used in various domestic cigarette enterprises. However, this equipment does not have the function of inspecting the appearance quality of cigarettes during the filter tipping process. It only relies on manual visual inspection for inspection, and problems such as missed inspections and false inspections of cigarette defects often occur. Quality defects such as uneven paper length and uneven pasting affect the appearance...

AI Technical Summary

Problems solved by technology

Although the online detection device uses machine vision to detect the quality of cigarettes, it does not use different technical processing for different defective cigarettes and does not consider interference factors such as noise and provide specific image processing and analysis methods. , therefore, the detection effect and detection speed cannot be expected

Images