Advanced Computer Vision's in Veterinary (Dogs & Cats) Research Using TRACOR

The research paper proposes an innovative approach to provide computers with a vision to effectively recognizing whether animals present in video recordings are cats or dogs using deep learning techniques, notably Convolutional Neural Networks (CNNs). Unique feature combines RGB and thermal imaging as the data sources for accurate computer vision. Thermal imaging helps in heat signatures captures, whereas RGB imaging captures actual visible details, these details increasing the proposed algorithmic accuracy in terms of animal identification. Obstacles like fur, cages, and other occlusions are handled using multi-stage algorithm. During the initial phase, intuitive preprocessing pipeline is in place to handle barriers then improvise RGB and thermal image quality. Once initial preprocessing effectively overcoming obstacles next phase is all about handling the accuracy in a variety of environmental situations about the animal visualization through extensive experiments done on a diverse dataset.