Identification of a Zebra Based On Its Stripes through Pattern Recognition

Visual animal biometrics are commonly used because they provide a non-invasive and cost-effective means of wildlife tracking. Cameras that are motion triggered capture nearby images of wild animals in their natural environments. This collection of chaotic photos taken in the field was used to create a biometric database of individual animals that can be distinguished from their coat markings. The SIFT Descriptors and FLANN parameters were used in this paper to simplify the dataset by extracting only the patterned parts of the body that can be used for resemblance detection. It was tested using two factors: colour match and feature match.