Sorting with AI

Brief description of the project

In our innovative research to optical sorting, advanced technology meets practical application. Our research focuses on the optimization of sorting systems through the use of Artificial Intelligence (AI), in particular Convolutional Neural Networks (CNNs) and Transformer, which outperform conventional image processing algorithms. CNNs show remarkable potential, especially for materials with complex visual properties and high occupancy densities. They close the gap for materials that can be distinguished with the trained eye, but do not allow simple differentiation based on color or shape. They work much faster and in some cases even more accurately than a manual sorter.

Project goals

Traditionally, sensor-based sorting techniques have been based on color and shape analysis or simple statistical models.  AI methods are revolutionizing this field and offer significant improvements in object classification and recognition. Studies on AI-based image analysis have been demonstrating the potential for sensor-based sorting for several years now, but practical implementation has been lacking. Our Research & Development is breaking new ground by integrating the latest AI methods into practical sorting systems. The AI locates and classifies objects in the camera images. Shortly after the camera’s line of sight, pneumatic valves are precisely activated to remove unwanted objects from the material stream. Thanks to efficient implementation and high-performance graphics cards, the sorting system is designed for real-time operation—without any cloud computing. Even at high material throughput and belt speeds exceeding 3 m/s, a response time of less than 30 ms is achieved. 

Fields of application and evaluation

The effectiveness of our approach has already been demonstrated in two challenging sorting tasks: the sorting of minerals with a high dust content and the detection of foreign bodies in food. These applications show the versatility and robustness of our AI-based sorting system. The technology can be used not only for applications such as food quality assurance, but also for material processing in waste recycling or mineral refining. We invite you to explore our comprehensive study that evaluates the real-time capabilities of the latest AI methods in sensor-based sorting, integrates advanced deep learning models and assesses sorting quality in practical applications. Discover how our work bridges the gap between theoretical research and practical implementation and sets a new standard in optical sorting technology. Read our publication, the download can be found at the bottom left of the web page.

Structure of the AI-supported sorting system

Sorting system for sorting construction waste with the following components: (1) Vibratory feeder (2) Conveyor belt (3) Camera box (4) Camera line (5) Lighting (6) Pneumatic valves.

AI-supported sorting in recycling

AI-supported sorting is used for material processing in the recycling sector. In this picture, concrete, porous concrete and bricks are separated.

AI prediction for construction waste

A high dust density in construction waste images makes it difficult for conventional algorithms to function correctly. The CNN reliably classifies the foreground objects and separates them from the background of the image.

AI-supported sorting in the food industry

In food quality assurance, foreign bodies are separated from the good material, to increase the purity of the good material, as here in tea sorting. This is achieved using a combination of a color camera, AI for data evaluation and compressed air nozzles to eject the foreign bodies.

AI prediction for foreign bodies in tea

The CNN makes a prediction of the materials. Despite similar color distributions, the foreign bodies can be reliably distinguished from the tea.

AI Predictions for Wood-Based Materials

For high-quality wood-based materials made from recycled wood, fiber length is a critical factor and depends heavily on the original type of wood or wood-based material. For this reason, materials are sorted into different categories of wood-based materials and solid wood. The AI identifies the type of wood-based material based on surface and structural characteristics.