machine vision

Demographic change, skills shortages, and stagnating productivity are threatening the competitiveness of German industry. At the same time, AI and digital assistance systems are opening up new opportunities: they make work more efficient and support skilled workers. But while they have long been part of everyday life, their potential in industry remains largely untapped—this is where this issue comes in with innovative concepts.

In contrast to the use of robots in standardized production processes, robots must be flexible and adaptable within dynamic logistics processes in order to cope with variable environmental conditions and non-standardized goods. Due to the recent advances in the field of artificial intelligence and networking through industry 4.0, robots will perform complex tasks in logistics in a reliable way in future. A crucial component of a robot system represents the interpretation of the work environment with the help of multi-modal sensor systems, especially image processing systems. This paper describes applications for robotic systems in logistics as well as a concrete example of focusing on the interpretation of multi-modal sensor data for the automation of a logistics task.

The major components of a machine vision system are image acquisition and image processing. The complexity of image acquisition is based on the huge number of degrees of freedom. This article introduced an extension of a sensor simulation tool which enables a user to simulate the current sensor positions and field of views. The results are registered point clouds from different sensor positions merged to a final object point cloud. Now it is possible to evaluate image processing techniques based on the simulated 3D data, without explicit results from the real sensor behaviour. The benefits of the tool were demonstrated in a real micro cold forming scenario. Future work focuses on additional sensor noise models. Adding these models results in a more realistic sensor simulation framework.

Due to the increased product miniaturization, a number of new applications and market opportunities open up for mechanical micro-manufacturing. In the manufacturing process with part dimensions less than one millimeter and tolerances in the micrometer range occur so-called “size effects”. These prevent a simple scaling of processes known methods from the macro level and lead to an increased occurrence of quality deviations. In conclusion, the process capability according to ISO 21747 is affected and therefore the application of statistical process control (SPC) is more difficult. In this paper, the interaction between technical and logistical quality objectives in terms of logistical quality control are analyzed at the example of micro cold forming. Thereby, methods of statistical process control and fuzzy control are used.

Due to the increasing miniaturization in all areas, mechanical manufacturing processes of micro-components become more important. The combination of high production cycles and low manu-facturing tolerances in the micrometer range requires a comprehensive quality management, which seeks efficient processes with low quality costs. Because of the small component sizes and the difficulties associated with the handling process, the classic visual inspection retires as testing procedure. In combination with a customized micro-manufacturing surface metrology, an efficient image processing system is needed to identify surface imperfections such as cracks, dents and scratches. This paper presents different standards and applications on the basis of quality management that affect the surface inspection of micro components. By reference to a micro-thermoformed component from the Collaborative Research Centre (CRC) 747, the prototypical implementation of an automated image processing system is ...