1. Fast defect detection for large scale photoelectric devices utilizing compressive sensing
- Author
-
Kai Xie, Liu Yan, and Quan Lei
- Subjects
Engineering ,Compressed sensing ,Scale (ratio) ,business.industry ,Feature (computer vision) ,Component (UML) ,Fast scanning ,Electronic engineering ,Photoelectric effect ,business ,Sample (graphics) ,Structured light - Abstract
Conventional scanning method for photoelectric devices is slow and ineffective, especially in large scale applications. Motivated by the fact that the failed or defected components are sparse, this paper proposed a fast scanning method utilizing compressive sensing. The proposed method measures the total output of the sample device under pattern controlled structured light beams. With a few numbers of photo-electric current measurements, the defected components are located in the compressive sensing manner. The concept of propose method and the detection probabilities of the proposed method are evaluated. Simulation results indicating that the sparse defects can be compressively located using a few measurements, and the detection speed can be accelerated by 10 times when the ratio of defects is less than 1.5%; this detection probability is sufficient for real application. The additional advantages of proposed detection system are that no mechanical movement component is required, and the testing for each single cell is no longer required, since the system measures the total output of the whole panel. This feature brings great benefits to the applications for testing the devices that have been assembled or installed.
- Published
- 2017
- Full Text
- View/download PDF