As industrial machines enter the 'computational era', the application of industrial AI in industrial visual quality inspection is becoming increasingly crucial. WeLinkirt's 2D AI AOI equipment provides an efficient solution for logo/silk-screen printing inspection on product bodies in the consumer goods industry.
User Scenario: On the production line of a leading consumer product manufacturer, after the logo/silk-screen printing process on the product body, quality inspection of the product's logo and silk-screen is required. The products include various consumer electronic products and daily necessities, and the inspection objects are planar elements such as logo patterns and text silk-screen on the product surface.
Pain Points: In the 'computational era' of industrial machines, traditional inspection methods face many dilemmas. Manual inspection is inefficient, with only about 200 products inspected per hour, and the missed-detection rate is as high as 3%. It is easy to miss the judgment of minor defects. At the same time, the false-alarm rate reaches 15%, resulting in a large number of products needing re-inspection, which increases labor costs and time costs. In addition, when the product type is changed, the re-debugging time of traditional inspection equipment is as long as 30 minutes, seriously affecting the production rhythm.
Technical Principle
WeLinkirt's 2D AI AOI equipment uses high-resolution 2D imaging technology, which can clearly capture the subtle features of logos and silk-screens. Its hardware is equipped with high-precision industrial cameras, which can achieve micron-level image acquisition. In terms of algorithms, it uses deep-learning secondary image-judging technology. By learning from a large number of positive samples, an accurate defect-recognition model is constructed. The deep-learning algorithm has powerful feature-extraction and classification capabilities, which can accurately distinguish normal patterns from defective patterns, effectively avoiding misjudgments caused by factors such as image noise and uneven illumination. At the same time, the semantic false-alarm filtering function further optimizes the inspection results. Based on the understanding of the defect semantics, it eliminates false-alarm information that does not conform to the actual defect features, improving the accuracy of the inspection.
- High - resolution 2D imaging: High - precision industrial cameras collect micron-level images to ensure that small defects are clearly visible.
- Deep - learning secondary image-judging: Build a model by learning from positive samples to accurately identify various defects.
- Semantic false-alarm filtering: Eliminate false false-alarm information based on the understanding of defect semantics.
WeLinkirt Solution and Product
Centered around the 2D AI AOI equipment, this device has the ability of high-speed online full inspection and can complete the inspection while the products quickly pass through the production line. Its micron-level inspection accuracy can detect tiny flaws in logos and silk-screens. Combined with the DaoAI AI AOI software system, it can achieve 0-code automatic programming for a good product in 5 minutes. The APDT positive-sample/few-sample learning function only needs 1-20 good products to complete model training, greatly reducing the product-change time. On the production line, the 2D AI AOI equipment is installed after the logo/silk-screen printing process. When the products pass through the equipment on the conveyor belt, the equipment automatically collects images and conducts inspections, and the inspection results are feedback in real-time.
The 2D AI AOI equipment provides an efficient and accurate solution for logo silk-screen inspection of consumer products with its advanced technology.
Quantitative Results: After introducing the 2D AI AOI equipment, the inspection efficiency has been greatly improved, with about 800 products inspected per hour. The detection rate reaches 99.2%, and the missed-detection rate is reduced to <0.8%, effectively preventing defective products from entering the market. At the same time, the false-alarm rate is reduced by -65%, greatly reducing the re-inspection workload. The product-change time is shortened from 30 minutes to 5 minutes, improving the production flexibility and efficiency.
FAQ
What types of logo/silk-screen defects can the 2D AI AOI equipment detect?
The equipment can detect planar defects such as surface scratches, uneven ink, missing characters, and pattern offsets. It uses high-resolution 2D imaging and deep-learning algorithms to achieve high-precision inspection.
How long does it take to change the product type for the equipment?
Combined with the DaoAI AI AOI software system, 0-code automatic programming can be completed in 5 minutes. The APDT positive-sample/few-sample learning function makes product-change more efficient.
To what extent can the false-alarm rate of the equipment be reduced?
Through the semantic false-alarm filtering function, the false-alarm rate can be reduced by -65%, effectively reducing the re-inspection workload and improving the inspection efficiency.