Design And Implement Of A Compression Algorithm For Image Region Of Interest Based On JPEG

Automatic optical inspection (AOI) is generally adopted as electron components defect detection method in the Surface Mounted Technology (SMT) for the advantages of fast testing and high trapping rate due to use the technology of machine vision as the testing standard. As miniaturization and integration is a tendency for the circuit boards, image date size sharp increases to detect with high accuracy. The PC's processing speed and data transmission efficiency limited by the continuous processing architecture of current software, image hosting and storage , reduced the AOI's test speed severely, particularly in the system in which multiple cameras take and process image at the same time .In order to improve the processing speed and accuracy of AOI, a region of interest image compression algorithm based on JPEG compression standard is presented in this thesis. This algorithm processes different parts of the same image using different precision according to the different degrees of interest and decode every region separately. This way it improves the image compression ration and decoding efficiency greatly. The effect of this algorithm is verified with an AOI camera embedded this compression algorithm. The results show that the data transfer speed between the camera and the PC and test speed of AOI are greatly improved, which meet the requirements of precision and speed of current SMT excellently.