Image Classification Using CNN With Multi-Core and Many-Core Architecture

2021 ◽  
pp. 233-266
Author(s):  
Debajit Datta ◽  
Saira Banu Jamalmohammed
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Classification ◽  
Medical Sciences ◽  
Classification Techniques ◽  
The Past ◽  
Distributed Framework ◽  
Multiple Cores ◽  
Many Core

Image classification is a widely discussed topic in this era. It covers a vivid range of application domains like from garbage classification applications to advanced fields of medical sciences. There have been several research works that have been done in the past and are also currently under research for coming up with better-optimized image classification techniques. However, the process of image classification turns out to be time-consuming. This work deals with the widely accepted FashionMNIST (modified national institute of standards and technology database) dataset, having a set of sixty thousand images for training a model and another popular dataset of MNIST for handwritten numbers. The work compares several convolutional neural network (CNN) models and aims in parallelizing them using a distributed framework that is provided by the python library, RAY. The parallelization has been achieved over the multiple cores of CPU and many cores of GPU. The work also shows that the overall accuracy of the system is not affected by the parallelization.

scholarly journals Analysis of Convolutional Neural Network based Image Classification Techniques

2021 ◽  
Vol 3 (2) ◽  
pp. 100-117
Author(s):  
Milan Tripathi
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Classification ◽  
Rural Areas ◽  
Processing Time ◽  
Rapid Urbanization ◽  
Classification Techniques ◽  
Bar Code ◽  
Supermarket Industry

With the rapid urbanization and people moving from rural areas to urban time has become a very huge commodity. As a result of this change in people's lifestyles, there is a growing need for speed and efficiency. In the supermarket industry, item identification and billing are generally done manually, which takes a lot of time and effort. The lack of a bar code on the fruit products slows down the processing time. Before beginning the billing process, the seller may need to weigh the items in order to update the barcode, or the biller may need to input the item's name manually. This doubles the effort and also consumes a significant amount of time. As a result, several convolutional neural network-based classifiers are proposed to identify the fruits by visualizing via the camera for establishing a quick billing procedure in order to overcome this difficulty. The best model among the suggested models is capable of classifying pictures with start-of-art accuracy, which is superior than that of previously published studies.

scholarly journals Image Classification using Supervised Convolutional Neural Network

2019 ◽  
Vol 8 (2) ◽  
pp. 4505-4507
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Classification ◽  
Large Scale ◽  
Computational Time ◽  
The Past ◽  
Input Layer ◽  
Classification Tasks ◽  
High Processing ◽  
Model Algorithm

Deep learning algorithms, in particular Convolutional Neural Networks have made notable accomplishments in many large-scale image classification tasks in the past decade. In this paper, image classification is performed using Supervised Convolutional Neural Network (SCNN). In supervised learning model, algorithm learns on a labeled dataset. SCNN architecture is built with 15 layers viz, input layer, 9 middle layers and 5 final layers. Two datasets of different sizes are tested on SCNN framework on single CPU. With CIFAR10 dataset of 60000 images the network yielded an accuracy of 73% taking high processing time, while for 3000 images taken from MIO-TCD dataset resulted 96% accuracy with less computational time

Sign in / Sign up

Export Citation Format

Share Document