Home / Current Issue / Paper 1703672
Automated Image Captioning for Visually Impaired
Subject area: Science,Engineering and Technology · Area of research: Machine Learning
Abstract
The study of machine learning algorithms and processes that are suited for each image and language process. The use of existing packages in the implementation of machine learning algorithms. Implementation of an Associate in Nursing algorithmic programme that takes a photograph and explains it in full phrases. A list of domain experience and prospective models was developed after the demand analysis. Following that, a study of the available technologies that may aid with the creation of the appliance was conducted, and the look, The solution's implementation and validation began. A deep convolutional neural network is used to extract features and transfer learning is combined with a repeating neural network for description generation in this study. The implementation is done with Keras with a TensorFlow backend. It was possible to obtain Photo exploitation language is described by a trained model.
Keywords
image captioning, machine learning, model.
How to cite this paper
@article{1703672,
author = {Pankaj, Dr. Dayanand J, Yuvraj, Sangamesh Patil, Siddalinga},
title = {Automated Image Captioning for Visually Impaired},
journal = {Iconic Research And Engineering Journals},
year = {2022},
volume = {6},
number = {1},
pages = {432-440},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1703672.pdf},
abstract = {The study of machine learning algorithms and processes that are suited for each image and language process. The use of existing packages in the implementation of machine learning algorithms. Implementation of an Associate in Nursing algorithmic programme that takes a photograph and explains it in full phrases. A list of domain experience and prospective models was developed after the demand analysis. Following that, a study of the available technologies that may aid with the creation of the appliance was conducted, and the look, The solution's implementation and validation began. A deep convolutional neural network is used to extract features and transfer learning is combined with a repeating neural network for description generation in this study. The implementation is done with Keras with a TensorFlow backend. It was possible to obtain Photo exploitation language is described by a trained model.},
keywords = {image captioning, machine learning, model.},
month = {July},
}