A real case that occurs in the field that many images are obtained from CCTV which need to identify a match with someone's face. This study aims to provide a solution the problem of detecting the image object match. SIFT algorithm to complete object match detection because this method is invariant to changes in scale, rotation, translation, and science. The conditions that affect the difference in object fit values are the light around object and the presence or absence of blur the image. The results of the SIFT method test are that level of compatibility of the suspect's image with the image of the same person produces a high matching value compared to a different image design. The level of accuracy object match analysis gets a value 100%. It can be concluded that the SIFT method can be implemented to detect the match of suspect image objects and image designs.
Copyrights © 2022