ANIMAL GAIT IDENTIFICATION USING A DEEP LEARNING METHOD

Assessment of animal posture is important for understanding animal behavior and injury. Previously, research has concentrated on consolidate animal-trackable beacon with high-pace cameras to support walk detection, which considered as time wasting and usually unpractical approach. This paper explains the achievability of utilizing deep learning for markerless motion tracking. We have trained a neural network to find the animal's body parts and test the nature of the animal's walk based on the different movements and bodies of the animals, including quadrupeds, bipeds, and reptiles. Various animal exposure assessment methods have been investigated using two modern convolutional network models. A model by CNN neural network was used to understand this behavior. Deep learning (DL) is a part of ML that include various techniques. Various models of deep learning have been explored to identify horse body parts. Trajectory was used to analyze the horse's movement to determine if the horse was lame. We present these areas below, Convolutional Neural Networks (CNN), a deep learning algorithm in image classification processing and the most important classification learning method