Dog Identification using Soft Biometrics and Neural Networks

TL;DR

This paper explores dog identification using deep neural networks combined with soft biometrics like breed, height, and gender, achieving over 90% accuracy in breed classification and improving identification rates.

Contribution

It introduces a transfer learning-based neural network that fuses soft and hard biometrics for enhanced dog identification accuracy.

Findings

Achieved over 90% accuracy in breed classification.

Using soft biometrics increased dog identification accuracy from 78.09% to 84.94%.

The proposed method outperforms traditional face-based identification methods.

Abstract

This paper addresses the problem of biometric identification of animals, specifically dogs. We apply advanced machine learning models such as deep neural network on the photographs of pets in order to determine the pet identity. In this paper, we explore the possibility of using different types of "soft" biometrics, such as breed, height, or gender, in fusion with "hard" biometrics such as photographs of the pet's face. We apply the principle of transfer learning on different Convolutional Neural Networks, in order to create a network designed specifically for breed classification. The proposed network is able to achieve an accuracy of 90.80% and 91.29% when differentiating between the two dog breeds, for two different datasets. Without the use of "soft" biometrics, the identification rate of dogs is 78.09% but by using a decision network to incorporate "soft" biometrics, the identification rate can achieve an accuracy of 84.94%.