Adaptive Image Segmentation by Combining Photometric Invariant Region and Edge Information

ÐAn adaptive image segmentation scheme is proposed employing the Delaunay triangulation for image splitting. The tessellation grid of the Delaunay triangulation is adapted to the semantics of the image data by combining region and edge information. To achieve robustness against imaging conditions (e.g., shading, shadows, illumination, and highlights), photometric invariant similarity measures, and edge computation is proposed. Experimental results on synthetic and real images show that the segmentation method is robust to edge orientation, partially weak object boundaries, and noisy, but homogeneous regions. Furthermore, the method is robust to a large degree to varying imaging conditions. Index TermsÐImage segmentation, adaptive splitting, integrating region and edge information, photometric color invariance, noise robustness.