Automatic Acquisition For Transparent Images And Automatic Position For Defects Of Optical Fiber Panel

Optical fiber panel is a new photonic devices raised in recent years. As the character of high resolution and light-gathering power, it has been the indispensable constituent part of optoelectronic imaging devices and systems. However, some defects exit in optical fiber panel because various Uncontrollable factors affect the optical fiber panel production, so it has great significance to detect the defects.As the extensive use of optical fiber panel, the performance request is also higher, so it's very necessary and urgent to detect the defects quantitatively, quickly and accurately. And this subject is just to do partial research about the detection.By research on the features of defects in optical fiber panel, basing on the various defects detection methods, the way using detection edge of diaphanous images to test the defects. It's determined to use improved Canny Operator to detect the defects edge by massive experiments. And then, the devices of the corresponding collection system are purchased, assembled and debugged based on research all elements and determination of the final automatic collection method and process for diaphanous images. Experiments show that the collection system is feasible. Finally, the collected images are further optimized by the Morphology method in connection with the ultimate pictures of diaphanous images, and the results of detection for diaphanous images edge of optical fiber panel are obtained to realize automatically locating.The system devised in this paper can achieve automatically collecting diaphanous images and locating defects, and it has higher detection accuracy and stronger adaptation ability, and it also has some application value to provide scientific theoretical basis for improving the images transmission quality of optical fiber panel.