REINFORCED CONTRAST ADAPTATION

被引：6

作者：

Tizhoosh, Hamid R. ^{[1
]}

Taylor, Graham W. ^{[1
]}

机构：

[1] Univ Waterloo, Dept Syst Design Engn, PAMI Res Grp, Waterloo, ON N2L 3G1, Canada

来源：

INTERNATIONAL JOURNAL OF IMAGE AND GRAPHICS | 2006年 / 6卷 / 03期

关键词：

Reinforcement learning; Q-learning; image enhancement; contrast; subjectivity;

D O I：

10.1142/S0219467806002379

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

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.

引用

页码：377 / 392

页数：16

共 26 条

[1] Quality measurement and use of pre-processing in image compression [J].

Algazi, VR ;

Avadhanam, N ;

Estes, RR .

SIGNAL PROCESSING, 1998, 70 (03) :215-229

[2] Subjective evaluation of MPEG-4 video codec proposals: Methodological approach and test procedures [J].

Alpert, T ;

Baroncini, V ;

Choi, D ;

Contin, L ;

Koenen, R ;

Pereira, F ;

Peterson, H .

SIGNAL PROCESSING-IMAGE COMMUNICATION, 1997, 9 (04) :305-325

[3] Statistical evaluation of image quality measures [J].

Avcibas, I ;

Sankur, B ;

Sayood, K .

JOURNAL OF ELECTRONIC IMAGING, 2002, 11 (02) :206-223

[4]

Betts B. J., 1998, TECHNICAL REPORT

[5] Adaptive target recognition [J].

Bhanu, B ;

Lin, YQ ;

Jones, G ;

Peng, J .

MACHINE VISION AND APPLICATIONS, 2000, 11 (06) :289-299

[6] Scenarios and design cognition [J].

Carroll, JM .

IEEE JOINT INTERNATIONAL CONFERENCE ON REQUIREMENTS ENGINEERING, PROCEEDINGS, 2002, :3-5

[7] The subjective image quality of direct digital and conventional panoramic radiography [J].

Gijbels F. ;

De Meyer A.-M. ;

Bou Serhal C. ;

Van den Bossche C. ;

Declerck J. ;

Persoons M. ;

Jacobs R. .

Clinical Oral Investigations, 2000, 4 (3) :162-167

[8] A computational approach to image quality [J].

Janssen, TJWM ;

Blommaert, FJJ .

DISPLAYS, 2000, 21 (04) :129-142

[9] Image quality and dose in computed tomography [J].

Jurik, AG ;

Jessen, KA ;

Hansen, J .

EUROPEAN RADIOLOGY, 1997, 7 (01) :77-81

[10] Reinforcement learning: A survey [J].

Kaelbling, LP ;

Littman, ML ;

Moore, AW .

JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 1996, 4 :237-285

← 1 2 3 →