Our client wanted to maintain an effective inventory management system through scanning and tracking every item of their inventory. With thousands of items arriving at a time, this nature of monitoring was difficult to set up and maintain with traditional practices, especially if you want to precisely identify product damages, inspect the quality and store the data efficiently. ThinkPalm developed an efficient object detection framework and GUI application for windows by leveraging deep learning capabilities. This solution utilized complex vision tasks such as image tagging, quality inspection, segmentation, or object detection and could be customized to identify specific defects or abnormalities in products. The framework applies object detection at a pixel-by-pixel level and is capable of being implemented into any of the client systems.
Identify/localization of defects with their corresponding label (welding defect, scratch, crack etc.) which are available with databases.
Quality inspection in manufacturing any products viz. automobile, textile, electronics, etc.
A Deep learned trained model can be customized according to the defect to be identified. Purely deep learning approach with digital image processing techniques.
Our applications visual inspection solution helped the client gain robust end-to-end workflow support for various industrial objects deep learning approaches with digital image processing techniques. The enterprise-grade application provided a complete ecosystem to label identify and localize defects with their corresponding label for seamlessly identifying, processing, and categorizing based on different types of defects. Our application helped the client classify each object precisely, evaluating it under the diversity of appearances, illumination conditions and backgrounds. The client was able to easily implement the application as it provided a user-friendly graphical user interface (GUI) and did not require expert programming skills to operate.
Our client strived to deliver the highest quality during every stage of the production or assembly process, ensuring that each object has the right shape or color or texture, and are free from any blemishes. By implementing a deep learning assisted quality inspection framework, our application significantly improved inspection cycle time and inspection accuracy. This enabled their teams to efficiently scale model run times and simplify the inference processes for quality inspection in manufacturing. Our client was able to efficiently reduce manufacturing errors as they continuously adjusted to market demand, product updates and new product rollouts.
Our object detection solution made computer vision more accessible and simplified the process of training, deploying, running and managing models. By leveraging the capabilities of image processing, our client was able to efficiently and accurately inspect whether the objects had any type of defects. Once the defect is identified, the model can be trained and customized with deep learning, so that similar defects can be easily identified. This intuitive solution is simple to operate, requires minimal operator training, and scales globally with the use of ubiquitous mobile hardware and Windows OS. ThinkPalm’s object detection solution is engineered to continuously improve accuracy with data management and training hardware. It ensured that our clients utilized a scalable foundation for real-time computer vision quality inspection and identified potential quality issues in real-time.
SKYLOGIQ is a manufacturer specializing in inspection and measurement-based in Hamamatsu City, Shizuoka Prefecture. Since their founding, they have consistently developed inspection and measurement systems centered on light and images. With their immense expertise in the latest technologies such as AI and IoT, they have developed innovative solutions that help clients gain a competitive advantage and most importantly, delivered continuous value. In the Artificial Intelligence domain, SKYLOGIQ has a lineup of Deep Sky that enables flexible recognition by deep learning, and in the IoT field, they have a lineup of Easy Monitoring that processes images from dozens of network cameras.
Contact us and we'll have one of our experts reach out to you and discuss how we can lead your project to success.