Comparison of Local Binary Pattern and Local Ternary Pattern for Classification of Arthritis in Knee X-Ray Images Using Euclidean Distance Method

Arthritis is the most common inflammation in bonejoints and this progressive disease often leads to early disability and joint deformities. By the early diagnosis and treatment of the Arthritis, the damage to the joins can be reduced. A number of therapeutic approaches are now widely available for the diagnosis of this disease. Imaging of the affected joints plays a vital role in the diagnosis. In this paper, a novel classification system for the classification of OA in knee x-ray images based on Local Binary Pattern (LBP) and Local ternary pattern(LTP) is presented. The classification is achieved by extracting the histograms of LBP and LTP of the knee x-ray image. Then classifier system based on K Nearest Neighbor (KNN) is constructed. This system classifies the knee x-ray images into normal or abnormal, and the abnormal severity into medium or worst cases. 50 knee x-ray images are used to evaluate the proposed system. The classification rate achieved is very satisfied. Keywords— Osteoarthritis, Knee X ray images, Local binary pattern, Local ternary pattern