本研究針對FPC(Flexible Printed Circuit)連接器之端子接腳,利用機器視覺技術發展一套檢測系統。在連接器的製造過程中,常因為端子接腳與塑膠外殼在組合時發生端子接腳之歪斜、斷裂等瑕疵,造成產品品質下降。目前業界大多是以人工檢測之方式對連接器做瑕疵檢測,這樣的方式不僅相當耗時,且檢測一致性偏低。在本研究中,係針對這些問題,發展一套檢測技術來取代人工檢測,期望達到降低生產成本、提高檢測效率及減少人力資源浪費。 在研究中之檢測系統整合了光學與數位影像處理技術,將待測物投打近同軸正向光源與背向光源,以雙攝影機擷取待測物正面與側面之圖像,再由數位影像處理技術分析圖像,判定OK/NG。此外,本研究提出了一種基於灰階強度之分佈結構為基礎之圖樣搜尋演算方法,不但可以快速且準確的搜尋到端子接腳區域,且大大提升檢測系統之適用性與便利性。
This thesis develops an optical inspection system for FPC (Flexible Printed Circuit) connectors on machine vision. During the inserting process, pins of the connector might be broken and curved. It will reduce the quality of products. Now, most of inspection factory heavily depend on human vision. This way raises the inspection time and does not reach the quality standard. In this research, we develop an optical inspection system to replace human vision and expect to increase the inspection efficiently, the quality of products and the quantity of output. In this research, the inspection system integrates optic with image processing. First, the top view and the side view of object image which is lighted up by directional front lighting and directional back lighting are captured by two CCD cameras. Then process the image data and analyse by digital image processing technique to decide OK/NG. Besides, a pattern searching algorithm which based on gray level structure of template is presented. This method can find locations of pin fast and increase convenience and power of inspection system.