A method to detect calibration patterns using fast adaptive binarization and plane homography for images having complicated backgrounds and taken under lighting condition of industrial fields is proposed. The preprocessing step of involving the calibration of a camera, as required to measure object dimensions, must be able to extract calibration points from a calibration pattern. However, proper lighting conditions for the camera calibration of a measurement system are rarely provided in industrial settings. In this paper, a plane homography of calibration points extracted by fast adaptive image binarization is used for the automatic detection of a local set of calibration points. The integral image method is used as a central part for the rapid processing of adaptive image binarization. The extracted point set provides a transformation matrix of the plane homography and guides the extraction of other calibration points for cluttered backgrounds with irregular lighting conditions. During the calibration of the camera, a calibration point filtering procedure by a data In/Out system iteratively selects more correct calibration points while rejecting outliers and noisy data. Experiments using actual images verify that the proposed method can be applied to camera calibration with poor quality images having cluttered backgrounds obtained under uneven illumination.