An important issue in industrial quality control is the inspection of rapidly moving surfaces for small defects such as scratches, dents, grooves, or chatter marks. This paper investigates the potential of the EyeRIS 1.3 camera as a state-of-the-art camera based on “cellular neural networks” (CNN) for this application in comparison to conventional image processing systems. Based on experimental data from an aluminum wire drawing process where defects with a lateral size of 100 μm have to be detected at feeding rates of 10 m/s, the potential specifications for other surface inspection applications are estimated. Using the relation between the lateral defect size and the feeding rate as a figure of merit, the CNN based system outperforms conventional image processing systems by an order or magnitude in this particular application. In general, the lighting system limits the performance at lower defect sizes and the computational power at larger defect sizes and fields of view.

Today image processing using a coaxial camera setup is used to monitor the quality of laser material processes such as laser welding, cutting, ablation or scribing. For real-time control of highly dynamic laser processes these systems are far to slow. This article proposes a sensing system for the next step: Using image based quality features in a real-time algorithm to form an instant feedback signal with up to 14 kHz in order to maintain the process in the desired state. The key component of the control system is a camera based on Cellular Neural Networks (CNN). This Single Instruction Multiple Data technology enables real time image processing by integrating processing units in every camera pixel. Moreover each of the pixel units is interconnected with its nearest neighbors which is optimal for most image processing algorithms. This article describes the CNN technology together with results obtained for CNN-based feedback systems designed for the closed-loop control laser ablation and laser welding algorithms.

CNN based cameras are able to process images in the kHz range in machine vision systems, as demonstrated for the closed loop control of laser welding processes. Other applications do not benefit from a comparable light source. This paper reports the current state of a LED based lighting for the inspection of cylindrical metallic surfaces on aluminum wires. Frame rates of 10 kHz and exposure times of 20 ¿s have been achieved so far.