In the production of automotive parts, the quality of brake discs is of crucial importance. WeLinkirt (DaoAI) provides an efficient and accurate solution for detecting scratches on brake discs with its advanced AI vision technology.
User Scenario: On the brake disc production line of a leading automotive parts manufacturer, the main product is automotive brake discs, and the detection object is the scratch defects on the surface of the brake discs. As a key component of the automotive braking system, the surface quality of the brake disc directly affects the driving safety of the vehicle. During the production process, the brake disc may be scratched for various reasons, so it needs to be strictly inspected.
Pain Points: The traditional detection method mainly relies on manual visual inspection, which is not only inefficient but also has a relatively high rate of missed detections and false alarms. According to statistics, the missed detection rate of manual inspection is as high as 10%, and the false alarm rate also reaches 8%. In addition, manual inspection also faces problems such as high labor costs and long change-over times. When changing to a different model of brake disc, the detection standards and processes need to be readjusted, and the change-over time is as long as 30 minutes, which seriously affects the production efficiency. At the same time, the scratches on the brake disc vary in shape, size, depth, and length, which also increases the difficulty of detection.
Technical Principles
WeLinkirt (DaoAI) uses advanced AI vision algorithms and imaging technologies to solve the problem of brake disc scratch detection. In terms of algorithms, deep learning algorithms are used to train a large number of brake disc scratch samples, enabling the model to learn the characteristics of different types of scratches. Deep learning algorithms have strong feature extraction capabilities and can accurately identify the location and characteristics of scratches from complex images. In terms of imaging, a self-developed 3D camera is used to photograph the brake disc to obtain its three-dimensional morphology information. The 3D camera can provide more abundant surface information and can clearly detect some hidden scratches. By combining 3D morphology reconstruction technology with deep learning algorithms, the depth and severity of scratches can be more accurately judged.
- Deep learning algorithms can automatically learn the characteristics of scratches, avoiding the limitations of traditional algorithms that require manual feature design, and improving the accuracy and generalization ability of detection.
- The three-dimensional morphology information obtained by the 3D camera can provide more surface details and can more accurately detect tiny scratches and depth information. In contrast, traditional 2D cameras can only obtain planar images, which are prone to missing some key information.
- 3D morphology reconstruction technology can accurately restore the surface morphology of the brake disc. Combined with deep learning algorithms, the depth and severity of scratches can be quantitatively analyzed, providing a more accurate basis for subsequent processing.
WeLinkirt (DaoAI) Solutions and Product Introduction
WeLinkirt (DaoAI) provides the DaoAI AI AOI software system and the DaoAI 2D / 3D AI AOI equipment to solve the problem of brake disc scratch detection. The DaoAI AI AOI software system has powerful feature recognition capabilities. Through the visual basic model, it can achieve 0-code automatic programming within 5 minutes for a single good product. At the same time, the system supports APDT positive sample / few-sample learning, and only 1-20 good products are needed for model training, greatly reducing the workload of sample collection. In addition, the system also has a semantic false-alarm filtering function, which can effectively reduce the false-alarm rate. The DaoAI 2D / 3D AI AOI equipment uses a self-developed 3D camera and 3D morphology reconstruction technology, which can detect the hidden solder joints, coplanarity, and micrometer-level morphology of the brake disc, making the detection of scratches on the brake disc surface more accurate and comprehensive. During the implementation process, the equipment is installed on the brake disc production line, and the software system analyzes and processes the collected images to achieve real-time detection of brake disc scratches.
WeLinkirt's (DaoAI) solutions provide efficient and accurate guarantees for automotive brake disc scratch detection through advanced technologies and products.
Quantitative Results: After adopting WeLinkirt's (DaoAI) solutions, the detection rate of brake disc scratches has increased to 98%, and the missed detection rate has been reduced to less than 2%. The false-alarm rate has been reduced by -70%, greatly reducing unnecessary re-inspection work. The change-over time has been shortened from the original 30 minutes to 5 minutes, improving the production efficiency. These results indicate that WeLinkirt's (DaoAI) AI vision solutions have significant advantages in automotive brake disc scratch detection.