Modeling Deep into Learning: A framework-based approach towards image detection

Abstract

Purpose: Detecting Specie is a popular task performed in the field of biodiversity. With the innovation in technology, specie detection can be done using deep learning. As we all know that Deep Learning is a field of machine learning having huge set of layers of neural network that train the model very deeply. In Image classification, every layer in deep learning classifies image using one feature. The authors propose a framework to detect species using the deep learning techniques.

Design/Methodology/Approach: Literature is reviewed to understand the application of deep learning techniques. Then Android platform is used to detect species using the deep learning techniques.

Findings: The framework would scan the specie, identify it and give information about the identified specie. For identification purpose, inception model from Google is retrained using TensorFlow

Paper Type: Theme based purely and emphasize on the application of deep learning techniques