Deep Learning Pet Identification Using Face and Body

According to the American Humane Association, millions of cats and dogs are lost yearly. Only a few thousand them found returned home. In this work, we use deep learning help expedite procedure finding dogs, for which new dataset is collected. We applied transfer methods on different convolutional neural networks species classification animal identification. The framework consists seven sequential layers: data preprocessing, classification, face body detection with landmark techniques, alignment, identification, soft biometrics, recommendation. achieved an accuracy 98.18% classification. identification layer, 80% was achieved. Body resulted in 81% accuracy. When using addition increased 86.5%, 100% chance that would be our top 10 recommendations matching. By incorporating animals’ biometric information, system can identify animals 92% confidence.