Abstract
Automatic Optical Inspection (AOI) is introduced in the manufacturing process. Detected defect is classified by the human eys check and human eye check may cause problem of unbalanced accuracy and that of cost. Based on these reasons, automatic defect classification is desired to the manufacuturing process. This paper proposes a convolutional neural network (CNN) of multiple input images with two different connection layers using two test images taken under two different conditions of illumination. Comparison is demonstrated in the experiments and the result suggests that better accuracy is obtained from the multi-input CNN which connects the two different connection layers near input layer. The performance of the proposed approach was validated with the obtained result of experiments.
Highlights
Printed Circuit Board (PCB) is crucial part of electronic device it needs to be properly investigated before get launched
Paper [3] improves the accuracy of classification using two images taken under the different Lighting conditions, and detects the defect region from the difference between the test image and the reference image which is prepared in advance
It is confirmed that side main lighting images gives better accuracy than coaxial main images from Table 3 and Figure 10. These results suggest that CNN2 gives the higher classification accuracy by learning the features between two different lighting images at around input layer
Summary
Printed Circuit Board (PCB) is crucial part of electronic device it needs to be properly investigated before get launched. Automatic inspection systems are used for this purpose but due to more complexity in circuits, PCB inspections are more problematic. This problem leads to new challenges in developing advanced automatic visual inspection systems for PCB. Automatic Optical Inspection (AOI) has been commonly used to inspect defects in printed circuit board during the manufacturing process. An AOI system generally uses meth-ods which detects the defects by scanning the PCB board and analyzing it. AOI uses methods like Local Feature matching, image Skeletonization and morphological image comparison to detect defects and has been very successful in detecting defects in most of the cases but production problems like oxidation, dust, contamination and poor reflecting materials leads to most inevitable false alarms. To reduce the false alarms is the concern of this paper
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
