Edge Detection To Detect Broken Bone On X-Ray Images Using Kirsch Method

Mhd Furqan (Department of Computer Science, Faculty of Science and Technology State Islamic University of North Sumatera Medan)
Rakhmat Kurniawan (Department of Computer Science, Faculty of Science and Technology State Islamic University of North Sumatera Medan)
Abdul Halim Hasugian (Department of Computer Science, Faculty of Science and Technology State Islamic University of North Sumatera Medan)
Reza Muhammad (Department of Computer Science, Faculty of Science and Technology State Islamic University of North Sumatera Medan)

INFOKUM
Vol. 10 No. 1 (2021): Desember, Data Mining, Image Processing, and artificial intelligence

Bone is the most important part of the human body and not a few people who have broken bones caused by accidents, neglect and because of age. One of the methods used to identify broken bones is to read the x-ray images manually. Manual checking takes longer and allows mistakes in making decisions when doing a manual check. What's more in reading an x-ray image requires a strong background light (as lighting) to make objects in the X-ray image appear clearer, so we need a method that can facilitate the orthopedic doctor in identifying broken bones. The method proposed in this study is edge detection using the kirsch method. Edge detection aims to improve the appearance of margins, boundaries or objects in the image. The Kirsch method will detect the edges of the eight cardinal directions, namely east, northeast, north, west, northwest, southwest, south and southeast, by convoluting the image using eight kernels. From the results of testing with these methods each cardinal direction has different results. In this study the best results from the eight cardinal directions are the southeast.