Research progress of food packaging defect detection based on machine vision

Abstract

Food packaging can develop defects during the production process due to various factors. The types of packaging defects are numerous with complex background. Machine vision detection, which uses visual imaging and computer information processing to complete tasks such as identification, detection, and measurement of packaging, has faster execution speed and higher accuracy compared to traditional manual inspection. This can significantly improve the degree of production automation. This article analyzes the common defects in food packaging and their causes, introduces traditional machine vision detection algorithms, and explores the research application of deep learning algorithms in food packaging defect detection. It also analyzes the prospects and challenges of applying detection algorithms in food packaging defect detection.

Keywords

• food packaging
• machine vision
• robotization
• deep learning
• defect detection