# A Technical Report on Image Classification using AWS

Source: [A Technical Report on Image Classification using AWS](<https://www.sota2.com/research/paper/68fa6f2e1e209131cea444b3>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T01:15:37.285Z

Authors: Balakrishna Phani Kommanaboina, Aditya Goverdhana, Naveena Kanderi, Jagadeesh Karri

Published: 2023-05-02T17:46:39.000Z

## Abstract

This project aims to build an elastic web application that can automatically scale out and scale in on-demand and cost-effectively by utilizing cloud resources, specifically from Amazon Web Services (AWS). The application is an image classification service exposed as a RESTful web service for clients to access. The infrastructure is divided into two tiers, the Web-Tier and the Application-Tier, with the former providing a user interface for uploading images, while the latter contains core functionality for image classification, business logic, and database manipulation functions. AWS EC2, SQS, and S3 resources are utilized to create this infrastructure, and scaling in and out of resources is determined by the number of incoming images. The project successfully demonstrated the implementation of an image classification application using AWS, which can be used in various industries, such as medical diagnosis, agriculture, retail, security, environmental monitoring, and manufacturing. However, the evaluation of the system based on metrics of response time, boot time, and accuracy highlighted some issues that need to be addressed to improve the application's performance. Overall, the scalability and cost-effectiveness of the infrastructure make it a suitable choice for developing image classification applications.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2305.01634>)

[PDF](<https://arxiv.org/pdf/2305.01634.pdf>)