Home // International Journal On Advances in Software, volume 13, numbers 1 and 2, 2020 // View article


Improve Operations of Real-Time Image Classification Utilizing Machine Learning and Knowledge Evolution

Authors:
David Prairie
Paul Fortier

Keywords: Machine Learning; Tensor Flow; Image Classification

Abstract:
This paper delves into the generation and use of image classification models in a real-time environment utilizing machine learning. The ImageAI framework was used to generate a list of models from a set of training images and also for classifying new images using the generated models. Through this paper, previous research projects and industry programs are analyzed for design and operation. The basic implementation results in models that classify new images correctly the majority of the time with a high level of confidence. However, almost a quarter of the time the models classify images incorrectly. This paper attempts to improve the classification accuracy and improve the operational efficiency of the overall system as well.

Pages: 69 to 79

Copyright: Copyright (c) to authors, 2020. Used with permission.

Publication date: June 30, 2020

Published in: journal

ISSN: 1942-2628