<![CDATA[The Indonesian Robot Contest (KRI) is an annual event organized by the Ministry of Research, Technology, and Higher Education. One of its divisions, the Indonesian Search and Rescue Robot Contest (KRSRI), focuses on designing robots for disaster victim identification. This research aims to develop a victim identification system for the KRSRI robot of ITN Malang using image processing techniques and the K-Nearest Neighbors (KNN) algorithm, designed to distinguish between actual victim dolls and dummy dolls based on color and shape characteristics. The research utilizes a Raspberry Pi 3 Model B+ and a Raspi Cam V2 camera to capture images, which are processed using image segmentation techniques to detect the orange-colored victim doll, while the KNN algorithm is applied for object classification. Experimental results show that the accuracy of victim detection varies depending on the K parameter in the KNN algorithm and the distance between the camera and the object. At a distance of 10 cm, 15 cm, and 20 cm, the highest accuracy achieved is 100% with certain K values, whereas at 25 cm, the accuracy drops significantly, reaching only 43.75% to 31.25%.]]>
