Sprayed character online visual detection method based on convolutional neural network

Abstract

The invention provides a sprayed character online visual detection method based on convolutional neural network. The sprayed character online visual detection method comprises the steps of dividing characters in an image under an offline status, classifying the characters, constructing a character repertoire, and performing training through an improved convolutional neural network learning method to form a stable classifier; shooting pictures, dividing the characters and classifying the characters in real time during online detection, and removing unqualified products. By means of the sprayed character online visual detection method, real-time performance is ensured while detection accuracy is improved greatly, and requirements for high real-time performance and accuracy of online detection process of the sprayed characters at bottoms of pop-top cans can be met.

Description

Technical field [0001] The invention relates to the field of on-line detection of the characters on the bottom of the can by using machine vision, in particular to an image processing method for the character recognition of the bottom of the can, which is applied to the industrial field and has high real-time requirements. Background technique [0002] With the rapid growth of my country's food and beverage industry, the demand for cans as containers is also increasing. The quality of cans often requires inspection and tracking of product information. At present, major manufacturers often use can bottom inkjet characters to track product information, but it is inevitable that there will be defects such as missing characters or partial incomplete characters and chaotic characters during the coding process. Therefore, how to print codes in real time Character recognition and detection to eliminate substandard products in time is an urgent problem to be solved. [0003] The current r...

AI Technical Summary

Problems solved by technology

At present, major manufacturers mostly use the coded characters on the bottom of cans to track product information, but it is inevitable that there will be defects such as missing characters, incomplete characters, and confusing characters during the coding process. Therefore, how to spray codes in real time Character recognition and detection, in order to remove unqualified products in time is an urgent problem to be solved

[0003] The current research on the online detection technology of inkjet characters mainly focuses on two aspects. One is the research on the preprocessing method. In order to meet the high real-time requirements of recognition, the preprocessing process should be as simple and efficient as possible; the other is the feature extraction method Research. At present, the feature extraction methods are mostly traditional structural features and statistical features. These features are better for traditional printed character recognition, but for easy-to-break, easy-to-distort inkjet characters, the recognition effect is not good.

Images