On-line intermittent hollow filter stick detection method based on regional feature analysis

Abstract

The invention belongs to the technical field of machine vision detection, and provides an on-line intermittent hollow filter stick detection method based on regional feature analysis, which comprisesthe following steps: (1) performing color space conversion and multichannel analysis on a filter stick image acquired by imaging equipment of an on-line intermittent hollow filter stick detection system so as to improve the distinction degree of a bright section and a dark section of a filter stick; (2) dividing the extracted channel data with the highest distinction degree into a bright section area and a dark section area by adopting a maximum between-cluster variance method; (3) after a bright section area and a dark section area of the intermittent hollow filter stick are divided, calculating the phase and the length of the subsection, determining the phase of the subsection by using an ensemble learning algorithm, further calculating the length of the subsection, and judging whether the filter stick product is qualified or not according to the phase and the length of the subsection of the intermittent hollow filter stick. According to the method, the image characteristics of the intermittent hollow filter stick are extracted and analyzed, the corresponding detection algorithm is optimized, and the high-speed and high-quality production requirements are met.

Description

technical field [0001] The invention belongs to the technical field of machine vision detection, and in particular relates to an online intermittent hollow filter rod detection method based on regional feature analysis. Background technique [0002] The production process of cigarette filter rods will produce defective products, usually including segment phase, segment length defects, etc. The traditional detection method of defective hollow filter rods adopts the detection method of manual sampling, but this method has disadvantages such as strong subjectivity of detection effect, low efficiency, high false detection rate and missed detection rate, so it cannot meet the needs of current industrial production . Other common filter rod detection methods include sensor detection and microwave detection. These two methods basically meet the needs of industrial production, but still have the disadvantage of high detection error rate. In addition, these two detection methods hav...

AI Technical Summary

Problems solved by technology

The traditional detection method of defective hollow filter rods adopts the detection method of manual sampling, but this method has disadvantages such as strong subjectivity of detection effect, low efficiency, high false detection rate and missed detection rate, so it cannot meet the needs of current industrial production

Other common filter rod detection methods include sensor detection and microwave detection. These two methods basically meet the needs of industrial production, but still have the disadvantage of high detection error rate. In addition, these two detection methods have strict requirements on the physical characteristics of the product. In this way, the false detection rate of the system is further reduced

Images