Web-Based Application for Digital Image Classification using TensorFlow

Videv, Ivan and Draganov, Ivo (2018) Web-Based Application for Digital Image Classification using TensorFlow. In: 53rd International Scientific Conference on Information, Communication and Energy Systems and Technologies (ICEST 2018), June 28-30, 2018, Sozopol, Bulgaria.

[thumbnail of ICEST2018_30.pdf]
Preview
Text
ICEST2018_30.pdf

Download (547kB) | Preview

Abstract

In this paper, we investigate the capabilities of applying TensorFlow for digital image classification. A complete web-based application is created with easy portability to various operating systems, which cover all the functionalities incorporated into the library. It follows the client-server model and take advantage of all the options for tuning, such as random crop, random scaling, etc. Testing with a set of 772 images of 3 types of objects taken in natural environment prove the applicability of the framework for classification purposes. The system can be extended to solve general and strictly specialized tasks.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Digital Image Classification, TensorFlow, Deep Learning, Convolutional Neural Network
Subjects: Scientific Fields (Frascati) > Engineering and Technology > Electrical engineering, electronic engineering,information engineering
Divisions: Faculty of Technical Sciences
Depositing User: ICEST
Date Deposited: 18 Mar 2020 17:08
Last Modified: 18 Mar 2020 17:08
URI: https://eprints.uklo.edu.mk/id/eprint/4246

Actions (login required)

View Item View Item