Toothpaste tube shoulder defect detection method based on Gabor filter

Abstract

The invention discloses a toothpaste tube shoulder defect detection method based on a Gabor filter. The method includes the following steps: carrying out pixel size calibration on a camera of a visionmeasurement system to obtain calibration data; acquiring a front image of a toothpaste tube shoulder through the camera; carrying out region-of-interest extraction for the above-mentioned image, andobtaining a to-be-detected region by dividing; carrying out Gabor filtering on the to-be-detected region to obtain a result graph; and according to an area / perimeter feature value in the result graph,judging whether a defect exists. According to the method, interference of other parts of workpieces is avoided through preprocessing the front image of the toothpaste tube shoulder, and the calculation amount of subsequent algorithm operations is decreased; and the Gabor filter is adopted, sensibility on edges of the image is realized, very good direction and scale selection can be provided, parameters are easy to adjust, very good illumination adaptability is realized, and a detection recognition speed and an accuracy rate are high. The toothpaste tube shoulder defect detection method basedon the Gabor filter provided by the invention can be widely applied to the field of product quality detection.

Description

technical field [0001] The invention relates to the field of product quality detection, in particular to a method for detecting toothpaste tube shoulder defects based on a Gabor filter. Background technique [0002] The quality and safety of products has always been a key link in industrial production, which is directly related to the sales share of products in the market. Only high-quality products can gain a place in the market competition. With the continuous development of the manufacturing industry, in order to ensure the competitiveness of products, the demand for improved product quality and inspection speed is increasing. However, the quality inspection of most products today is still implemented by manual methods. Manual inspection is time-consuming and labor-intensive, and at the same time requires the company to invest a lot of human and financial resources to ensure the quality of the product. Manual detection requires workers to repeat similar tasks, which can...

AI Technical Summary

Problems solved by technology

[0004] At present, manual inspection is still the main method of final quality inspection of toothpaste tubes. Due to the structure of toothpaste tubes, this inspection method obviously has low work efficiency and poor detection reliability. Using toothpaste that is not up to standard may bring serious health problems to consumers. Hidden danger

[0005] In recent years, the surface inspection technology based on machine vision has made great progress, and some detection hotspots have appeared, but the defect detection for toothpaste tube shoulders has not been realized so far.

However, most surface defect detection algorithms still face problems such as low contrast between defect areas and non-defect areas, high similarity between noise and defects, slow detection speed and low recognition accuracy; at the same time, most algorithms are based on their respective surface defects. It is not suitable for the three-dimensional structure of the toothpaste tube shoulder, and cannot meet the requirements of its speed accuracy.

Images