Energy-efficient Image Classification on Edge Devices: Implementation and Evaluation

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2024 | Volume :11 | Issue : 03 | Page : –
By
vector

Shital Thakkar,

vector

Arpit Acharya,

vector

Richa Singh,

vector

Harsh Bhalala,

vector

Pinkesh Patel,

vector

Vinay Thumar,

  1. Associate Professor, Department of Electronics and Communication Engineering, Dharmsinh Desai University (DDU), Nadiad, Gujarat, India
  2. Student, Department of Electronics and Communication Engineering, Dharmsinh Desai University (DDU), Nadiad, Gujarat, India
  3. Student, Department of Electronics and Communication Engineering, Dharmsinh Desai University (DDU), Nadiad, Gujarat, India
  4. Student, Department of Electronics and Communication Engineering, Dharmsinh Desai University (DDU), Nadiad, Gujarat, India
  5. Student, Department of Electronics and Communication Engineering, Dharmsinh Desai University (DDU), Nadiad, Gujarat, India
  6. Student, Department of Electronics and Communication Engineering, Dharmsinh Desai University (DDU), Nadiad, Gujarat, India

Abstract document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_abs_107105’);});Edit Abstract & Keyword

Image classification is a computer vision problem where an algorithm determines class or label for a given image. Various real time applications like object recognition, medical diagnosis, person recognition, etc. Image classification property on edge devices is useful for autonomous vehicles, surveillance, and healthcare and IoT deployments. The advancement of deep learning-based methods and GPU devices allows efficient processing locally. The study utilizes a GPU-powered board, recognized for its strength in visual computing, to run a Convolutional Neural Network (CNN) classification algorithm. Deep learning with GPUs speeds up image analysis by extracting features and constructing neural networks, enabling efficient processing of both static and real-time data. The combination of edge device efficiency and GPU acceleration illustrates the potential for strong image classification in practical scenarios. We demonstrated classification tasks such as binary classification of cats versus dogs and multi-class classification on the CIFAR10 benchmark dataset.

Keywords: GPU, Deep Learning, Convolution Neural Networks (CNN), CIFAR10, computer vision,

[This article belongs to Journal of Image Processing & Pattern Recognition Progress (joipprp)]

How to cite this article:
Shital Thakkar, Arpit Acharya, Richa Singh, Harsh Bhalala, Pinkesh Patel, Vinay Thumar. Energy-efficient Image Classification on Edge Devices: Implementation and Evaluation. Journal of Image Processing & Pattern Recognition Progress. 2024; 11(03):-.
How to cite this URL:
Shital Thakkar, Arpit Acharya, Richa Singh, Harsh Bhalala, Pinkesh Patel, Vinay Thumar. Energy-efficient Image Classification on Edge Devices: Implementation and Evaluation. Journal of Image Processing & Pattern Recognition Progress. 2024; 11(03):-. Available from: https://journals.stmjournals.com/joipprp/article=2024/view=0

Full Text PDF

Full Text PDF

References
document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_ref_107105’);});Edit

  1. Aliady H, Utari DT. GPU based image classification using convolutional neural network chicken dishes classification. Int. J. Advance Soft Compu. Appl. 2018 Jul 1;10(2):1-3.
  2. Ang LM, Seng KP. GPU-based embedded intelligence architectures and applications. Electronics. 2021 Apr 16;10(8):952.
  3. Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Communications of the ACM. 2017 May 24;60(6):84-90.
  4. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. 2014 Sep 4.

 

  1. Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. InProceedings of the IEEE conference on computer vision and pattern recognition 2017 (pp. 4700-4708).
  2. Tan M. Efficientnet: Rethinking model scaling for convolutional neural networks. arXiv preprint arXiv:1905.11946. 2019.
  3. Kamron Bhavnagri. Convolutional Neural Networks: Basic Theory in a Nutshell. The Data Science Swiss Army Knife. 2019. Available from: https://www.kamwithk.com/convolutional-neural-networks-basic-theory-in-a-nutshell
  4. Davidson RL, Bridges CP. Error resilient GPU accelerated image processing for space applications. IEEE Transactions on Parallel and Distributed Systems. 2018 Mar 6;29(9):1990-2003.
  5. Gavali P, Banu JS. Deep convolutional neural network for image classification on CUDA platform. InDeep learning and parallel computing environment for bioengineering systems 2019 Jan 1 (pp. 99-122). Academic Press.
  6. NVIDIA DGX Platform. NVIDIA. 2024. Available from: https://www.nvidia.com/en-us/data-center/dgx-platform/ ‌
  7. CIFAR-10 and CIFAR-100 datasets. Toronto.edu. 2024. Available from: https://www.cs.toronto.edu/~kriz/cifar.html
Regular Issue Subscription Review Article
Volume 11
Issue 03
Received 14/08/2024
Accepted 26/09/2024
Published 11/10/2024
504

function myFunction2() {
var x = document.getElementById(“browsefigure”);
if (x.style.display === “block”) {
x.style.display = “none”;
}
else { x.style.display = “Block”; }
}
document.querySelector(“.prevBtn”).addEventListener(“click”, () => {
changeSlides(-1);
});
document.querySelector(“.nextBtn”).addEventListener(“click”, () => {
changeSlides(1);
});
var slideIndex = 1;
showSlides(slideIndex);
function changeSlides(n) {
showSlides((slideIndex += n));
}
function currentSlide(n) {
showSlides((slideIndex = n));
}
function showSlides(n) {
var i;
var slides = document.getElementsByClassName(“Slide”);
var dots = document.getElementsByClassName(“Navdot”);
if (n > slides.length) { slideIndex = 1; }
if (n (item.style.display = “none”));
Array.from(dots).forEach(
item => (item.className = item.className.replace(” selected”, “”))
);
slides[slideIndex – 1].style.display = “block”;
dots[slideIndex – 1].className += ” selected”;
}