Image enhancement plays a critical role in image processing. This paper introduces an integrated approach using three methods: Super-Resolution Generative Adversarial Net- work (SRGAN), Convolutional Autoencoder (CAE), and Zero- Reference Deep Curve Estimation (ZeroDCE), all embedded within a web application to provide image enhancement opportunities to users. SRGAN enhances image resolution by generating high-quality images from low-resolution inputs, with improved adjustments in Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) values. The Convolutional Autoencoder denoises impulse noise while preserving key image features. ZeroDCE automatically enhances low-light images by adjusting pixel values using Light-Enhancement curves (LE- curves), Deep Curve Estimation Network (DCE-Net), and Non- Reference loss functions. In conclusion, this paper contributes to the broader field of computer vision and image processing, providing academic and practical understanding of the performance and limitations of these three models, paving the way for further developments in the field. The code is available at https://github.com/robinson-pujara/Refinetograph.git.
목차
Abstract 1. Introduction 2. System Architecture 2.1. System Design 2.2. Working Model 3. Methodology 3.1. Overview 3.2. Data and Training 4. Results and Analysis 4.1. SRGAN Model Analysis 4.2. CAE Model Analysis 4.3. ZeroDCE Model Analysis 5. Discussion 6. Conclusion Acknowledgments References
키워드
CAEPSNRSRGANZeroDCESSIM
저자
Santosh Gaire Sharma [ Department of Electronics and Computer Engineering, Institute of Engineering, Advanced College of Engineering and Management ]
Robinson Pujara [ Department of Electronics and Computer Engineering, Institute of Engineering, Advanced College of Engineering and Management ]
Prabesh Aryal [ Department of Electronics and Computer Engineering, Institute of Engineering, Advanced College of Engineering and Management ]
Surendra Shrestha [ Faculty of Science, health and technology, Nepal open university, Nepal ]
Corresponding Author
한국AI디지털융합학회(구 한국디지털융합학회) [The Korean Academic Society of AI Digital Convergence]
설립연도
2015
분야
사회과학>경영학
소개
본 학회는 디지털 경영에 관련된 디지털 미디어, 디지털 통신, 디지털 방송, 디지털 콘텐츠, 디지털 문화, 디지털 사회, 디지털 유통, 디지털 금융, 디지털 물류, 디지털 정책, 디지털 기술, 디지털 교육 그리고 디지털과 아날로그의 비교 등에 대한 학제간 연구와 실사구시적인 적용을 통하여 디지털 경영의 발전과 한국이 세계적인 디지털 강국으로 성장하기 위한 학술적인 기반과 실무적인 지침을 조성하는 것을 목적으로 하고 있습니다.
간행물
간행물명
IJICTDC [International Journal of Information Communication Technology and Digital Convergence]