Combination of Different Technologies as the Basis of Surface Inspection
Vision Technology is not a single technology, but represents a combination of a number of different technologies that are implemented as needed. Vision Technology systems are based on the non-contact measurement of electromagnetic radiation (i.e. visible light, infrared light, X-rays) by a camera, automatic processing and evaluation of image data and the generation of a context-dependent decision by a system computer.
The abstract system function of a Vision Technology System is generally the control, regulation or monitoring of an automated process or the generation of (decision) data from an imaged scene.
Sequences of Vision Technology Systems
Every Vision Technology System arrives at a context-related decision through a sequence of three process steps:
- Image acquisition
- Image preprocessing
- Image evaluation
Image Acquisition as a Process Basis
In image acquisition a camera generates a digitized representation of the scene. Image acquisition is always specific to the object and task. The objective is as far as possible a perfect representation of the scene in respect of the desired system functionality.
Image acquisition is based on the disciplines of illumination technology, optics, camera technique and computer technology. In addition, it requires know-how in handling technology, IO-technique and the use of complementary sensors (e.g., in order to determine the presence of the inspection object).
Data Optimization in Image Preprocessing
In image preprocessing the image data is transformed to a more suitable format for image evaluation. In high-speed applications such as the surface inspection of glass in display industry, a drastic reduction in the volume of data delivered by the camera is often desirable, in order to ensure complete inspection of the paper despite computing capacity limitations.
As for image acquisition, there is no specific method for image preprocessing. The available methods are selected according to the application. Depending on the system functions, one or more methods may be combined. Image preprocessing is based on computer, signal processing and software technologies.
As for Surface Inspection, Basler is offering an optimized, in-house developed hardware, which is well adapted to high scan rates. This hardware is able to process image data of several cameras in parallel.
Analysis of Image Data in Image Evaluation
In image evaluation, the preprocessed image data are analyzed, and certain actions are initiated depending on the results.
The goal is to identify defects in the material and to classify them by fault. A critical or large defect or a high number of defects ultimately leads to rejection of the test object. Moreover, the knowledge acquired can be used for process optimization.
Meanwhile, Basler has over 20 years of experience in image processing and evaluation in the software development field. Beyond simple defect identification – adapted to the specific requirements – we assess and accordingly classify the defects that we find.
