形状特徴量に基づく検出対象領域の自動検出 : マイクロ接合部領域の自動検出に関する研究(第2報)

概要

論文の詳細を見る
To detect a specific region in an image is required for the vision system. In the detection of the micro-joints, detection method of the region where should detected in regions with the similar shape and consideration for the scattering of the size are very important. In this research, the shape expression method for the micro-joints in a computer and weighting value for each shape characteristic are discussed. It is shown that the evaluation method for the size of each shape characteristic based on the fuzzy set theory and the analytic hierarchy process method is effective. Furthermore, it is clarified that the synthetic evaluation used the weighting value calculated by the analytic hierarchy process method improve the ability of detection. At applying this automatic detection system to the detection of the QFP soldered joints on a printed circuit board, it is shown that the detection ratio is 100% and miss-detection is 0.
社団法人溶接学会の論文
1998-05-05