Matching Images On The Face Of A Buddha Statue Using The Scale Invariant Feature Transform (SIFT) Method

The condition of the statue in Indonesia at this time many are missing, especially the face of the statue. 504 Buddha statues in the Borobudur temple, there are 248 statues which are now in a headless condition. The condition of the lost statue face is due to being stolen, still buried in the area around the temple, or has changed hands to another party. To find the statue's head according to the face, a matching image is needed, one of which uses the sift method. Buddha face statue matching can be done based on the characteristics of hair, eyebrow shape, lips, head shape or attributes found on the head of the statue. The image matching method can describe an image with a specific keypoint using Scale Invariant Feature Transform (SIFT), where each keypoint has a gradient orientation (GO) and gradient magnitude (GM) which are processed into features in the image registration process. The purpose of this study is to match the face images of Buddha statues with different angles. This study aims to match the face image of the original statue with the image of the statue chosen from different angles. The results of this study are that the same image with different angles obtains 90% accuracy.