Abstract:
Visual animal biometrics is an emerging research field of computer vision, pattern recognition, and cognitive science. Recently, cattle recognition has played a significant role in understanding, controlling and the outbreak of critical diseases, vaccination, production management, traceability, and ownership assignment of a livestock animal. The traditional animal recognition methodologies, such as ear-tagging, freeze-branding, ear-tattoos, embedded microchips, ear tips or notches-based, and electrical-based marking approaches, have been applied to recognize individual livestock animal. However, standard animal recognition procedures are invasive. The performance of conventional methods is not good due to their vulnerability to losses, easy duplication, and fraud of embedded tag number. These are major security issues and challenges for the identification of cattle throughout the world. Visual animal biometric systems are gaining more proliferations due to widespread applications to recognize individual cattle based on their primary biometric muzzle point image characteristics. This paper aims to provide a comprehensive review of cattle recognition and tracking from non-biometric recognition approaches (classical animal recognition methods) to visual animal biometric systems using muzzle point image pattern along with measurements and interpretations based on current state-of-the-art methods. Moreover, this paper demonstrates the basic deployment of the animal biometric system to uniquely identify the animals using their biometric characteristics. This study can hopefully encourage new multidisciplinary researchers and scientists to provide excellent efforts for the designing and development of adequate algorithms for solving the classification and recognition problems. The literature review is followed by the presentation of references for more details, incorporating applications and current trends. © 2019, The National Academy of Sciences, India.
