2D AI AOI Equipment Enables Defect Detection in Food Packaging

In the food/agriculture industry, product quality is the key to an enterprise's survival and development. As an important part of the product, food packaging quality directly affects the product's quality and the consumer experience. Therefore, efficient and accurate packaging inspection is an urgent need in the industry. The emergence of WeLinkirt's 2D AI AOI equipment provides a perfect solution to this need.

Industry Background and User Scenarios: In today's food/agriculture industry, consumers' attention to product quality and safety is constantly increasing, which makes it particularly important for enterprises to control product quality. Food packaging is not only the external image of the product but also an important means to protect the product and extend its shelf-life. The packaging process of a leading food manufacturer's production line covers food packaging sealing, label pasting, and product coding. The inspection objects include various types of food packaging such as plastic packaging and carton packaging. The enterprise needs to ensure that the packaging is tightly sealed to prevent food spoilage, the labels are correctly pasted to ensure that consumers can accurately obtain product information, and the coding is clear and complete for product traceability and management.

In - depth Analysis of Pain Points: Why is it Difficult?

Traditional inspection methods have many problems in packaging sealing, label, and coding inspection. In terms of quantitative indicators, the miss-detection rate is about 3%, which means that 3 out of every 100 products with defects may enter the market. This not only affects the consumer experience but also damages the brand reputation. The false-alarm rate is as high as 25%, which requires enterprises to re-inspect a large number of misjudged products, increasing a large amount of labor and time costs. In terms of labor costs, due to the need for a large number of manual spot inspections, enterprises invest heavily in labor wages and training.

The root cause of the difficulty of traditional methods lies in their technical limitations. Traditional inspections mainly rely on manual experience and simple optical principles, making it difficult to adapt to complex and changeable packaging defects. For example, when detecting tiny sealing gaps or fuzzy coding, manual inspections are prone to visual fatigue and judgment errors. Moreover, with the rise of non - Transformer architecture technology, traditional methods lack efficient algorithm support, making it difficult to meet the requirements of more efficient and accurate inspections and unable to effectively process large amounts of image data and complex defect features.

Technical Principles

WeLinkirt's 2D AI AOI equipment uses high-resolution 2D imaging technology and a deep-learning secondary image-judging algorithm. High - resolution 2D imaging achieves micron-level image resolution through high-precision industrial cameras and optical systems, enabling clear capture of tiny defects on the packaging surface. It's like equipping inspection with a pair of 'all-seeing eyes', making it difficult for any subtle flaws to escape. The deep-learning secondary image-judging algorithm is based on a non - Transformer architecture. Through learning from a large number of positive samples and a small number of samples, it can quickly and accurately identify various defects on the packaging surface.

Compared with traditional methods, the non - Transformer architecture has obvious advantages in processing such planar image data. It avoids the complexity of the Transformer architecture when processing simple planar images, reducing training time and computational resource consumption. Traditional methods may take a lot of time for model training and have high requirements for hardware resources. In contrast, the non - Transformer architecture of WeLinkirt's equipment can analyze a large number of images in a short time and has relatively low requirements for hardware resources, greatly improving the inspection efficiency.

Typical Application Scenarios

• Packaging Sealing Inspection: In the packaging sealing process, the equipment captures images of the sealing area through high-resolution 2D imaging technology to detect problems such as loose seals and gaps. The difficulty lies in that some tiny gaps may be difficult to detect with the naked eye, requiring the equipment to have high-precision imaging and analysis capabilities.
• Label Pasting Inspection: For label pasting, the equipment checks whether the label position is correct, whether there are wrinkles, and whether the pasting is firm. The difficulty is that the label materials and colors are diverse, which may interfere with image recognition. The algorithm needs to adapt to different label features.
• Product Coding Inspection: It checks whether the coding is clear and complete and whether the characters are correct. The clarity and quality of the coding may be affected by factors such as the coding equipment and packaging materials. The equipment needs to accurately identify coding defects in various situations.
• Packaging Appearance Inspection: It checks whether there are scratches, stains, and other defects on the packaging surface. Due to the different textures and colors of the packaging surface, accurately distinguishing normal textures from defects requires the equipment to have powerful image analysis capabilities.

Implementation Case

A large-scale food production enterprise with a wide variety of products and huge daily output. Before introducing WeLinkirt's 2D AI AOI equipment, the enterprise used traditional inspection methods and faced problems such as high miss-detection rates, high false-alarm rates, and long change-over times. During the implementation process, WeLinkirt's technical team customized the installation and debugging of the equipment according to the enterprise's production line characteristics and requirements. After a period of trial operation and optimization, the equipment was officially put into use. Before the implementation, the enterprise's miss-detection rate was about 3%, the false-alarm rate was as high as 25%, and the change-over time took several hours. After implementation, the detection rate increased to 99.4%, the miss-detection rate decreased to <0.6%, the false-alarm rate decreased by -63%, and the change-over time was shortened from several hours to 5 minutes.

WeLinkirt's 2D AI AOI equipment brings an efficient and accurate solution to food packaging inspection, significantly improving the enterprise's production efficiency and product quality.

WeLinkirt's Solution and Products

Centered around the 2D AI AOI equipment and combined with the DaoAI AI AOI software system. The 2D AI AOI equipment has the ability of high-speed online full-inspection and can conduct real-time inspection of food packaging on the conveyor belt with a detection accuracy of up to the micron level. The DaoAI AI AOI software system uses APDT positive-sample/less-sample learning technology. With only 1-20 good samples, it can complete 0-code automatic programming in 5 minutes, greatly shortening the change-over time. At the same time, the semantic false-alarm filtering function of the system effectively reduces the false-alarm rate.

Quantitative Achievements: After using WeLinkirt's solution, the detection rate increased to 99.4%, the miss-detection rate decreased to <0.6%, effectively preventing defective products from entering the market. The false-alarm rate decreased by -63%, significantly reducing the re-inspection workload. The change-over time was shortened from several hours to 5 minutes, improving the flexibility and production efficiency of the production line.

This article was generated by AI. Customer cases are simulated scenarios based on real product capabilities and figures are illustrative; see product pages for official benchmarks.