Active Contour for Applying CT Segmentation on a Single Image to Surrounding Multiple Images

Abstract

There is a growing interest in using X-ray CT data of industrial products for reverse engineering and model-based design. In this process, the image processing operation, called segmentation, for separating assembled parts into each single one is required. Currently, manual segmentation is often required and it costs a huge amount of man-time. In this study, we introduce a method that uses Active Contour and deep learning to apply a man-operated single segmentation to surrounding similar images.