Convolutional Neural Network based Inception v3 Model for Animal Classification

Abstract: Identification of similar types of objects in image processing now become regular task, but when dissimilar objects come into picture it becomes quite complex. Even it become more complicated when not just we need to identify objects but also categorizes them into their specific classes. We are using machine learning technique to classify the animal and put them into specified classes. Inception-v3 is the open source algorithm made by Google and used for objects classification. In this paper, based on Inception-v3 model in TensorFlow platform, we use the transfer learning technology to retrain the animal category datasets, which can greatly improve the accuracy of animal classification.

Keywords: Classification, Inception-v3, image processing, TensorFlow.