Deep Learning-Based Architectures for Recognition of Cow Using Cow Nose Image Pattern
Rotimi-Williams Bello, Abdullah Zawawi Talib, Ahmad Sufril Azlan Mohamed
Abstract
Stacked denoising auto-encoder and deep belief network are proposed as methods of deep learning for cow nose image texture feature extraction, and for learning the extracted features for better representation. While stacked denoising auto-encoder is applied for encoding and decoding of the extracted features, a deep belief network is applied for learning the extracted features and representing the cow nose image in feature space. Stacked denoising auto-encoder and deep belief network help in animal biometrics. Biometrics emanated from computer vision and pattern recognition and it plays an important role in the automated animal registration and identification process. Using the visual attributes of cow, and for the fact that the existing visual feature extraction and representation methods are not capable of handling cow recognition; deep belief network and stacked denoising auto-encoder are proposed. An experiment performed under different conditions of identification indicated that deep belief network outshines other methods with approximately 98.99% accuracy. 4000 cow nose images from an existing database of 400 individual cows contribute to the community of research especially in the animal biometrics for identification of individual cow.