Bone Image Segmentation and Recognition Using Image Processing Techniques

This paper presents a fast approach for bone image analysis including segmentation and recognition. Segmentation is carried out by K-means clustering of the gray scale image, whereas the recognition phase is based on feature extraction and two-level statistical classication. The proposed approach has applications in medicine and veterinary anatomy studies, orthopedics, paleontology, and archaeology. Several image features, including geometric and moment invariants (regular and Zernike), are derived for recognition. The rst-level classication is used to distinguish different kinds of bone and the second-level to recognize the right animal the bone is belong to. Two-dimensional structures, namely clusterproperty and cluster-features matrices, have been employed to evaluate different bone’s characteristics. Experimental results at the rst-level recognition exhibit better performance of the geometric features compared to moment invariants and Zernike moments. On the other hand, Zernike moments showed supremacy in differential diagnosis at the second level to recognize animals.