Maulisa Oktiana, Mutia Salima, Muharratul Mina Rizky, Ahmad Jazlan
Chest X-ray imaging is widely used for diagnosing pulmonary diseases. However, suboptimal contrast and image noise can degrade image quality and limit deep learning models' ability to extract discriminative features. This paper proposes the integration of contrast stretching as an image enhancement technique to improve the performance of a MobileNetV2-based chest X-ray classification system. Experiments were conducted on 854 chest X-ray images collected from Dr. Zainoel Abidin Regional General Hospital, Banda Aceh, Indonesia, comprising three classes: Normal, Pneumonia, and COVID-19. The experimental results show that the proposed approach improves classification accuracy from 91% to 95%, with consistent gains in precision, recall, and F 1 Score. The findings suggest that contrast stretching improves image quality and supports more effective feature learning in the lightweight MobileNetV2 architecture, contributing to better classification performance while maintaining low computational complexity. © 2026 IEEE.
Universitas Syiah Kuala, Dept. of Electrical and Computer Engineering, Banda Aceh, Indonesia; International Islamic University Malaysia, Dept. of Mechatronic Engineering, Gombak, Malaysia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock