Optical Surface Inspection
The system solution for inspecting the surfaces and geometries of materials uses high-definition line scan cameras to detect even the smallest of defects. A topological inspection, image filtering and a comprehensive defect classification with over 100 features are used to precisely and reliably detect defects and minimize false rejections. The automated inspection is characterized by intuitive operation. The automatic inspection programs teach-in process utilizes available with Deep Learning.
The identification of defects on surfaces is a special challange. The defects on complex functional and aesthetic technical surfaces. Deep Learning offers a new solution for those difficulties. Because it works without pre-definition of the expected defects. And also without the configuration of an inspection program.
The thresholding level defines the amplitude level (= severity) of the defects, the shape of the threshold surface defines the type of defect
Appearance defect are detected in locations where the local curvature has exceeded a certain threshold
Advantage of the Deep Learning Surface Detect
The algorthims can be integrated in any machine vision software.
The tool can be deployed in a ROI using for example the size of the anomaly surface defect as qualification criterion.
No teaching in advance
No setting parameters,
As the algorithm adapts itself automatically to any random surface
Evaluation of the inspected surface in less than 50 ms on Core i3