Home   >   CSC-OpenAccess Library   >    Manuscript Information
Comparison of Deep Learning Algorithms with Different Activation Functions for Brightness Image Enhancement
M. Fatih Yilmaz, Bekir Karlik
Pages - 12 - 24     |    Revised - 30-11-2024     |    Published - 31-12-2024
Volume - 13   Issue - 2    |    Publication Date - December 2024  Table of Contents
MORE INFORMATION
KEYWORDS
Deep Learning, Image Enhancement, Activation Function, Loss.
ABSTRACT
The training and refinement of deep learning-based image enhancement models are vital to their advancement. Researchers have examined various optimization methods, including adversarial training, multi-task learning, and perceptual loss functions, to enhance the quality and consistency of the improved results. This research explores brightness image enhancement techniques utilizing Autoencoder and Convolutional Neural Networks (CNN). Furthermore, the most appropriate activation functions for both approaches have been analyzed. Within this framework, functions like sigmoid, tanh, Softmax, ReLU, and leaky ReLU have been evaluated. The results suggest that tanh and leaky-ReLU are the most efficient activation functions for autoencoder neural networks, whereas sigmoid and SoftMax are the most appropriate for CNNs in image enhancement applications.
Anwar, S., & Li, C. (2020, August 20). Diving deeper into underwater image enhancement: A survey. Elsevier BV, 89, 115978-115978.
Augustauskas, R., & Lipnickas, A. (2020). Improved pixel-level pavement-defect segmentation using a deep autoencoder. Sensors, 20(9), 2557.
Bhardwaj, N., Kaur, G., & Singh, P K. (2017, October 4). A Systematic Review on Image Enhancement Techniques. Springer Nature, 227-235.
Chakraborty, T., S, U R K., Naik, S M., Panja, M., & Manvitha, B. (2024, January 17). Ten years of generative adversarial nets (GANs): a survey of the state-of-the-art. IOP Publishing, 5(1), 011001-011001.
Coghlan, S., & Quinn, T P. (2023, May 10). Ethics of using artificial intelligence (AI) in veterinary medicine. Springer Nature.
DeVries, T., & Taylor, G W. (2017, January 1). Improved Regularization of Convolutional Neural Networks with Cutout. Cornell University.
Dosovitskiy, A., & Brox, T. (2016). Generating images with perceptual similarity metrics based on deep networks. Advances in neural information processing systems, 29.
Gupta, R., Sharma, A., & Kumar, A. (2020, January 1). Super-resolution using GANs for Medical Imaging. Elsevier BV, 173, 28-35.
Hernández-García, Á., & König, P. (2018, January 1). Further Advantages of Data Augmentation on Convolutional Neural Networks. Springer Science+Business Media, 95-103.
Hu, L., Qin, M., Zhang, F., Du, Z., & Liu, R. (2021). RSCNN: A CNN-Based Method to Enhance Low-Light Remote-Sensing Images. Remote Sensing, 13(1), 62.
Karlık, B.; Aydın, S.; Kılıç, İ. (1996). Comparing Artificial Neural Networks (ANN) Image Compression Technique with Different Image Compression Techniques. Math. Comput. Applications, 1, 165-171.
Karlik, B. (2000). Comparison of widely used compression techniques for medical image. Proceedings of the PRASA, 105-110.
Karlik, B. (2006 June 29). Medical image compression by using vector quantization neural network (VQNN). Neural Network World, 16 (4), 341-348.
Khan, S., Ullah, I., Khan, F., Lee, Y., & Ullah, S. (2023, April 14). Historical Text Image Enhancement Using Image Scaling and Generative Adversarial Networks. Multidisciplinary Digital Publishing Institute, 23(8), 4003-4003.
Khosla, C., & Saini, B S. (2020). Enhancing Performance of Deep Learning Models with different Data Augmentation Techniques: A Survey. nternational Conference on Intelligent Engineering and Management (ICIEM), London, UK, June 1, 2020, 79-85, doi: 10.1109/ICIEM48762.2020.9160048.
Kirsten, L N., Piccoli, R., & Ribani, R. (2021, August 16). Evaluating deep neural networks for image document enhancement. https://doi.org/10.48550/arXiv.2106.15286
Kocyigit, Y., Korurek, M., & Karlik, B. (1999). ECG Data Compression by Artificial Neural Networks. Proceeding of the ELECO, 99, 338-340.
Long, J D., Shelhamer, E., & Darrell, T. (2014, January 1). Fully Convolutional Networks for Semantic Segmentation. Cornell University.https://doi.org/10.48550/arXiv.1411.4038
Mushtaq, Z. B., Nasti, S. M., Verma, C., Raboaca, M. S., Kumar, N., & Nasti, S. J. (2022). Super-resolution for noisy images using convolutional neural networks. Mathematics, 10(5), 777.
Mustafa, W A., Yazid, H., Mustafa, W A., Jamlos, M A., Zunaidi, I., Razlan, Z M., & Shahriman, A B. (2019, June 1). Image Enhancement Based on Discrete Cosine Transforms (DCT) and Discrete Wavelet Transform (DWT): A Review. IOP Publishing, 557(1), 012027-012027.
O’Mahony, N., Campbell, S., Carvalho, A., Harapanahalli, S., Hernandez, G. V., Krpalkova, L., & Walsh, J. (2020). Deep learning vs. traditional computer vision. In Advances in Computer Vision: Proceedings of the 2019 Computer Vision Conference (CVC), Volume 1 1 (pp. 128-144). Springer International Publishing.
Peng, Y., Sonka, M., & Chen, D Z. (2023, January 1). U-Net v2: Rethinking the Skip Connections of U-Net for Medical Image Segmentation. Cornell University. https://doi.org/10.48550/arXiv.2311.17791
Samek, W., Wiegand, T., & Müller, K. (2017, January 1). Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models. Cornell University.
Taylor, L., & Nitschke, G. (2017, January 1). Improving Deep Learning using Generic Data Augmentation. Cornell University. https://doi.org/10.48550/arXiv.1708.06020
Tensmeyer, C., & Martinez, T. (2017, November 1). Analysis of Convolutional Neural Networks for Document Image Classification. https://doi.org/10.48550/arXiv.1708.03273
Thomas, S A. (2021, January 1). Enhanced Transfer Learning Through Medical Imaging and Patient Demographic Data Fusion. Cornell University. https://doi.org/10.48550/arXiv.2111.14388
Tian, T., Wan, J., Song, Q., & Wei, Z. (2019). Clustering single-cell RNA-seq data with a model-based deep learning approach. Nature Machine Intelligence, 1(4), 191-198.
Zhang, Y., Tiňo, P., Leonardis, A., & Tang, K. (2021, August 24). A Survey on Neural Network Interpretability. Institute of Electrical and Electronics Engineers, 5(5), 726-742.
Mr. M. Fatih Yilmaz
Sheriff Latif LLC., Fremont, CA - United States of America
Professor Bekir Karlik
Computer Engineering, Epoka University Albania, Tirana-1000 - Albania


CREATE AUTHOR ACCOUNT
 
LAUNCH YOUR SPECIAL ISSUE
View all special issues >>
 
PUBLICATION VIDEOS