南久 List of Products and Services

Color sorting of glass bottles only using AI image processing. Other items such as empty cans and PET bottles are not sorted by color, only glass bottles are selected. By using HALCON, the deep learning training data consists only of the glass bottles we want to pick, and there is no need to train on prohibited items like PET bottles, empty cans, or cosmetic bottles! ⇒ Patent application planned. The image recognition accuracy for prohibited items is nearly 100%. Lighting does not require expensive options like commercially available white LEDs or UV lights. The learning method has been specialized, and the amount of image data used for training has decreased. Colors are sorted into brown, clear, green, etc. Only glass bottles of the specified color are picked by the robot using conveyor tracking. The robot can be flexibly modified to suit the end users, including: Collaborative robots, Parallel link robots, Articulated robots, and many other types. This sorting machine is a culmination of our company's technology, which knows everything about glass bottles! Plus, it uses the world-class image processing HALCON. Completion is just around the corner. Please look forward to it. *HALCON is a pioneering tool that allows for the construction of efficient image processing systems by freely combining over 2,000 high-functionality functions.

• Automatic sorting machine

Remove foreign substances contained in fluorescent lamp glass using an optical automatic sorting machine and an air gun. Only fluorescent lamp glass will be processed. Foreign substances will be recognized solely through high-speed image processing and removed with an air gun. Foreign substance removal rate: 99%.

• Automatic sorting machine
• Other inspection equipment and devices
• Glass

Using HALCON deep learning (global anomaly detection), we will automate the appearance inspection of automotive cast parts that are currently inspected visually.

• Visual Inspection Equipment
• Defect Inspection Equipment

Using HALCON deep learning object detection, we instantly AI-detect the classification of electronic components, their positions, and rotation angles. By combining this with robots, picking tasks become possible. It can be used for AI picking in food and automotive parts.

• Contract manufacturing

Using HALCON DL (Deep Learning) for image classification, as well as global anomaly detection, object detection, and other inspections, HALCON Deep Learning can be utilized depending on the inspection target. Of course, image processing is also possible based on rules. Note: HALCON is a pioneering tool that allows for the efficient construction of image processing systems by freely combining a vast array of over 2,000 high-functionality functions.

• Defect Inspection Equipment

Using a borescope, we will conduct an appearance inspection of automotive parts (cylindrical shapes with diameters of φ3 to φ6 mm) with HALCON DL (Deep Learning) to detect scratches, defects, dents, crimping, and dirt.

• Visual Inspection Equipment

AI image processing (deep learning) has been adopted for the glass recycling sorter. Initially, only high-quality glass is trained, and the AI extracts colored glass and identifies ceramics and stones through color sorting. Glass is allowed to free-fall from the feeder, and recognition of glass and ceramics is achieved through LED lighting transmission. There are 67 air guns, the feeder width is 1000mm, and the processing capacity is 3 tons per hour. The color sorting function can extract everything except transparent glass.

• Automatic sorting machine
• Glass
• Analytical Equipment and Devices

Wood chips and foreign objects are freely dropped, and only foreign objects are removed using deep learning and color sorting. Feeder width (foreign object identification range 1300mm) Device size: height 1500mm, width 1500mm, depth 1200mm It is a very compact device, making it easy to secure installation space. A video of the image processing is posted on YouTube: https://www.youtube.com/watch?v=Qa-gGEW5Bic https://www.youtube.com/watch?v=KSECGsqi_tw

• Automatic sorting machine

We can detect scratches, dents, and other imperfections on metal surfaces using high-performance AI image processing. Image processing is done using HALCON DL or HALCON. Cameras used are Basler and JAI UV cameras. If your company is facing challenges with appearance inspection, please feel free to provide samples, and we will conduct sample testing. By using HALCON, we can employ a wide range of detection methods compared to general image processing, making it possible to perform challenging appearance inspections that are difficult with standard image processing! Additionally, since AI image processing is also possible, we can handle appearance inspections that are difficult to detect consistently.

• Visual Inspection Equipment
• Contract manufacturing

We will pick only copper products of various shapes from the metal scraps refined by the shredder. Copper products come in various shapes, but we will accurately recognize their positions using AI image processing and pick them using conveyor tracking with a scara robot. Picking rate of over 90 percent.

• Automatic sorting machine

Foreign objects flowing on a high-speed conveyor (100 to 120 meters per minute) are recognized using AI image processing and removed with an air gun. The camera is a high-speed color camera, and the conveyor width is 500 millimeters.

• Automatic sorting machine

We detect and eliminate colorless transparent borosilicate glass (such as baby bottles and Pyrex) mixed in the cullet (mix, flint) using image processing. The light source for image processing utilizes UV light. We detect borosilicate glass flowing on the conveyor within the cullet. The detected borosilicate glass is picked by a robot. Currently, there has been no technology in the cullet recycling industry to eliminate colorless transparent borosilicate glass, but our company has developed this technology for the first time.

• Automatic sorting machine
• Glass
• Other inspection equipment and devices

We are currently sorting out foreign objects (stones, ceramics) mixed in the cullet by human eyes and hands, but we will automate the removal using a combination of AI-equipped image processing, vision systems, and conveyor tracking functions with scalar robots. This will definitely lead to a reduction in labor shortages, fixed costs, and an improvement in the quality of the cullet. There are no identical shapes, colors, or sizes for the ceramics and stones mixed in the cullet, but the AI-equipped image processing can reliably distinguish between good products and foreign objects, allowing the robot to pick only the foreign objects. While human visual inspection alone reduces reliability, this automatic sorting machine can consistently remove foreign objects. There is no doubt that the quality of the cullet will improve!! *This technology can also be applied to the removal of foreign objects in food products, not just cullet!!

• Automatic sorting machine
• Other inspection equipment and devices
• Glass

With a high-speed color line scan camera, when specifying a color from the mixed pellets, only that color pellet will be automatically sorted by AI. Using a high-speed line scan camera combined with high-speed processing AI image processing, only the pellets of the accurately specified color will be extracted. The pellets are allowed to free fall and are sorted at high speed using a color line scan camera. Processing capacity: 1 to 5 tons per hour Feeder width: 1000 millimeters The color to be extracted will be automatically selected by AI according to the customer's specifications.

• Automatic sorting machine
• Glass
• Other inspection equipment and devices

Using the vision system HALCON (image processing with AI functionality), we automatically recognize caps from cullet (broken glass fragments from bottles) and pick them using the conveyor tracking function of a SCARA robot. We will use a picking vacuum conveyor to suction the caps. Currently, caps are identified by human eyes and manually removed, but by introducing a robotic system, we can reduce labor costs and achieve stable cap removal! This will also lead to improved quality of the cullet.

• Automatic sorting machine
• Glass
• Image Processing Equipment

Using a high-speed, high-precision 8K line scan camera, the machine's structure is simple and free from failures, with high accuracy in foreign object removal, automatically eliminating foreign objects (ceramics, stones) within the cullet. It features low false reactions and a low defect rate. Cullet is allowed to free fall, and the identification of cullet and foreign objects is performed accurately in an instant using a light transmission method. A special LED bar lighting source is used to enhance the light transmission of the cullet, enabling stable foreign object detection. Our unique ejector control is implemented to improve foreign object removal accuracy. The color settings of the cullet and the size of foreign objects can be configured via a touch panel. We automate factories that currently rely on manual sorting of cullet, offering a lower price compared to other companies (overseas products) and achieving high foreign object removal accuracy, significantly improving the efficiency of cullet processing and enabling labor reduction. If there are individuals in the cullet processing industry considering automation, we encourage you to explore our solution. We have ample proven results and can confidently provide our services. Please feel free to discuss pricing as well. Cullet can be automatically processed into products without the need for manual sorting. Visits to Minami Kyu Co., Ltd. for demonstrations are also possible!

• Automatic sorting machine