| Along with the higher complexity and output of Printed Circuit Board(PCB) manufacture recent years, the traditional methods of PCB defect inspection such as got-up inspection can't meet demands of inspection gradually because of restricted contact, high cost or low efficiency. Therefore, the study of a automatic defect inspection system instead of manual work is meaningful in today's PCB production.According to optical inspection theory, the paper establishes a framework of defect detecting system based on image processing, constructs of the structure of software, researches on PCB's defect detecting algorithm in pivotal function modules of inspection software and application of algorithm, Which is implemented with VC++ program. The inspection system is composed of several parts, including scanner and computer image processing system. Inspection software of system includes modules that image acquisition, image pre-processing, threshold partition, image matching, image recognition.In software module, To give emphasis to researches and comes true all image process arithmetic involved in the system. In the model of image capture, a project method is used to extract the PCB image. In order to legible image, all kinds of method about image pre-processing are applied in the module of image pre-processing. In consideration of the problem of choosing threshold non-automatic, a new method of automatic choosing threshold based on pixel count is offered to choose six computing threshold arithmetic in the module of threshold partition. After matching the standard PCB image and defect PCB image by applying match methods based on character region, defects image could be acquired by XOR operation. In the module of image recognition, by analyzing the character of five defects, namely short circuit, open circuit, redundancy, notch, hollow, the software designs and implements arithmetic, and promotes defects information by window to convenient for worker resolving. |
PDF Full Text Request