In this paper, we proposed a method for measuring accurate positions and sizes of package and balls in a micro BGA. Also, we suggested four algorithms for detecting pixel defects and line defects of TFT-LCD panel. To find defects of BGA accurately, we focused on finding positions of package and ball...
In this paper, we proposed a method for measuring accurate positions and sizes of package and balls in a micro BGA. Also, we suggested four algorithms for detecting pixel defects and line defects of TFT-LCD panel. To find defects of BGA accurately, we focused on finding positions of package and balls. For fast operation, labeling method was applied on the * A thesis submitted to committee of the Graduate School of Hoseo University in partial fulfillment of the requirements for the degree of Master of Engineering in November 2005. input image. Then, we detected package and ball connected components using feature parameters respectively. After the detection of package component, we measured position and size of package by employing rectangular model which was constructed by the package information. After the detection of the ball components, we measured positions and radiuses of balls by employing circular models which were constructed by the ball informations. We did calibration based on four landmarks to measure real length, and we compared the measured results with the SEM data. Finally, we found that the accuracy of the proposed method is 94% in terms of ball’s radius. For higher accuracy in calibration, many landmarks in a wide range or restoration method based on dot grid is required. For TFT-LCD cell inspection, we separated the foreground and the background region for accurate detection of dark line defect, bright line defect, dark pixel defect, and bright pixel defect. We separated defect region by using morphology with the designed structure element which reflected the shape of TFT-LCD sub-pixels. For fast and accurate defects detection, we applied the morphology, followed by threshold and labeling. Also, we colored each defects for easier checking with the eye. The gray value of bright subpixel was normally 200. Then, we could detect the dark pixel defects with the gray value of 0 to 130. The gray value of dark sub-pixel was normally 19. Then, we could detect the bright pixel defects with the gray value of 59 to 255. By employing the proposed method, we could detect the defects of the 9 panels with the accuracy of 100%, in real time. Although the input images showed moire effects, we could detect defects using the proposed method. By examining the results of the experiments, we found that it is necessary to employ moire prevention method, for higher accuracy in subpixel defect detection. Further research in vision inspection algorithm will likely be the 3D vision inspection algorithm using two cameras for mura defect detection.
In this paper, we proposed a method for measuring accurate positions and sizes of package and balls in a micro BGA. Also, we suggested four algorithms for detecting pixel defects and line defects of TFT-LCD panel. To find defects of BGA accurately, we focused on finding positions of package and balls. For fast operation, labeling method was applied on the * A thesis submitted to committee of the Graduate School of Hoseo University in partial fulfillment of the requirements for the degree of Master of Engineering in November 2005. input image. Then, we detected package and ball connected components using feature parameters respectively. After the detection of package component, we measured position and size of package by employing rectangular model which was constructed by the package information. After the detection of the ball components, we measured positions and radiuses of balls by employing circular models which were constructed by the ball informations. We did calibration based on four landmarks to measure real length, and we compared the measured results with the SEM data. Finally, we found that the accuracy of the proposed method is 94% in terms of ball’s radius. For higher accuracy in calibration, many landmarks in a wide range or restoration method based on dot grid is required. For TFT-LCD cell inspection, we separated the foreground and the background region for accurate detection of dark line defect, bright line defect, dark pixel defect, and bright pixel defect. We separated defect region by using morphology with the designed structure element which reflected the shape of TFT-LCD sub-pixels. For fast and accurate defects detection, we applied the morphology, followed by threshold and labeling. Also, we colored each defects for easier checking with the eye. The gray value of bright subpixel was normally 200. Then, we could detect the dark pixel defects with the gray value of 0 to 130. The gray value of dark sub-pixel was normally 19. Then, we could detect the bright pixel defects with the gray value of 59 to 255. By employing the proposed method, we could detect the defects of the 9 panels with the accuracy of 100%, in real time. Although the input images showed moire effects, we could detect defects using the proposed method. By examining the results of the experiments, we found that it is necessary to employ moire prevention method, for higher accuracy in subpixel defect detection. Further research in vision inspection algorithm will likely be the 3D vision inspection algorithm using two cameras for mura defect detection.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.