REINFORCED CONTRAST ADAPTATION
- 1 July 2006
- journal article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Image and Graphics
- Vol. 6 (3), 377-392
- https://doi.org/10.1142/s0219467806002379
Abstract
Traditional image enhancement algorithms do not account for the subjective evaluation of human operators. Every observer has a different opinion of an ideally enhanced image. Automated Techniques for obtaining a subjectively ideal image enhancement are desirable, but currently do not exist. In this paper, we demonstrate that Reinforcement Learning is a potential method for solving this problem. We have developed an agent that uses the Q-learning algorithm. The agent modifies the contrast of an image with a simple linear point transformation based on the histogram of the image and feedback it receives from human observers. The results of several testing sessions have indicated that the agent performs well within a limited number of iterations.Keywords
This publication has 15 references indexed in Scilit:
- Observer-Dependent Image EnhancementPublished by Springer Science and Business Media LLC ,2003
- Reinforcement learning for rule extraction from a labeled datasetCognitive Systems Research, 2002
- A measure of quality for evaluating methods of segmentation and edge detectionPattern Recognition, 2001
- A computational approach to image qualityDisplays, 2000
- The subjective image quality of direct digital and conventional panoramic radiographyClinical Oral Investigations, 2000
- Quality measurement and use of pre-processing in image compressionSignal Processing, 1998
- Subjective image fidelity metric based on bit allocation of the human visual system in the DCT domainImage and Vision Computing, 1997
- Subjective evaluation of MPEG-4 video codec proposals: Methodological approach and test proceduresSignal Processing: Image Communication, 1997
- Subjective quality assessment of compressed imagesSignal Processing, 1997
- Image quality and dose in computed tomographyEuropean Radiology, 1997