A reinforcement agent for threshold fusion
Applied Soft ComputingPublished 10 January 2007
Maryam Shokri, Hamid R. Tizhoosh
Citations16
SJR quartileQ1
SJR score1.51
SNIP1.97
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A reinforced threshold fusion for image binarization is introduced which aggregates existing thresholding techniques and learns the optimal weights for different thresholds and segments the image globally.
Abstract
Finding an optimal threshold in order to segment digital images is a difficult task in image processing. Numerous approaches to image thresholding already exist in the literature. In this work, a reinforced threshold fusion for image binarization is introduced which aggregates existing thresholding techniques. The reinforcement agent learns the optimal weights for different thresholds and segments the image globally. A fuzzy reward function is employed to measure object similarities between the binarized image and the original gray-level image, and provide feedback to the agent. The experiments show that promising improvement can be obtained. Three well-established thresholding techniques are combined by the reinforcement agent and the results are compared using error measurements based on ground-truth images.
Keywords
Computer Science
Machine LearningQ-learning
8,970 Citations1992Christopher J. Watkins, Peter Dayan
This paper presents and proves in detail a convergence theorem forQ-learning based on that outlined in Watkins (1989), showing that Q-learning converges to the optimum action-values with probability 1 so long as all actions are repeatedly sampled in all states and the action- values are represented discretely.
Journal of Artificial Intelligence ResearchReinforcement Learning: A Survey
8,831 Citations1996Leslie Pack Kaelbling, Michael L. Littman +1 more
Central issues of reinforcement learning are discussed, including trading off exploration and exploitation, establishing the foundations of the field via Markov decision theory, learning from delayed reinforcement, constructing empirical models to accelerate learning, making use of generalization and hierarchy, and coping with hidden state.
MIT Press eBooksIntroduction to Reinforcement Learning
6,868 Citations1998Richard S. Sutton, Andrew G. Barto
In Reinforcement Learning, Richard Sutton and Andrew Barto provide a clear and simple account of the key ideas and algorithms of reinforcement learning.
Bell System Technical JournalError Detecting and Error Correcting Codes
5,086 Citations1950R. W. Hamming
The author was led to the study given in this paper from a consideration of large scale computing machines in which a large number of operations must be performed without a single error in the end result.
Journal of Electronic ImagingSurvey over image thresholding techniques and quantitative performance evaluation
4,514 Citations2004B. Sankur
40 selected thresholding methods from various categories are compared in the context of nondestructive testing applications as well as for document images, and the thresholding algorithms that perform uniformly better over nonde- structive testing and document image applications are identified.
Computer Vision Graphics and Image ProcessingA new method for gray-level picture thresholding using the entropy of the histogram
3,603 Citations1985J. N. Kapur, Prasanna K. Sahoo +1 more
Two methods of entropic thresholding proposed by Pun (Signal Process.,2, 1980, 223–237;Comput.16, 1981, 210–239) have been carefully and critically examined and a new method with a sound theoretical foundation is proposed.
ERROR DETECTING AND ERROR CORRECTING CODES
3,222 Citations2014Neha Chauhan, Pooja Yadav +1 more
In this paper all error detecting and error correcting mechanisms are studied and best mechanism on the basis of accuracy, complexity and power consumption is selected.
Computer Vision Graphics and Image ProcessingA survey of thresholding techniques
2,658 Citations1988Prasanna K. Sahoo, Soheil Soltani +1 more
This paper presents a survey of thresholding techniques and attempts to evaluate the performance of some automatic global thresholding methods using the criterion functions such as uniformity and shape measures.
Pattern RecognitionMinimum error thresholding
2,158 Citations1986Josef Kittler, J. Illingworth
A computationally efficient solution to the problem of minimum error thresholding is derived under the assumption of object and pixel grey level values being normally distributed and is applicable in multithreshold selection.
RPK (Politechniki Krakowskiej)Artificial Intelligence: A Modern Approach
1,746 Citations1995Stuart Russell, Peter Norvig
Pattern RecognitionImage thresholding by minimizing the measures of fuzziness
967 Citations1995Liang-Kai Huang, Mao‐Jiun J. Wang
A new image thresholding method based on minimizing the measures of fuzziness of an input image and a fuzzy range is defined to find the adequate threshold value within this range.
Computer Vision Graphics and Image ProcessingMoment-preserving thresolding: A new approach
908 Citations1985Wen‐Hsiang Tsai
Experimental results show that the new approach to automatic threshold selection using the moment-preserving principle can be employed to threshold a given picture into meaningful gray-level classes.
Computer Vision Graphics and Image ProcessingA comparative performance study of several global thresholding techniques for segmentation
570 Citations1990Sang Uk Lee, Seok Yoon Chung +1 more
Computer simulation results reveal that most algorithms perform consistently well on images with a bimodal histogram, however, all algorithms break down for a certain ratio of population of object and background pixels in an image, which in practice may arise quite frequently.
IEEE Transactions on Systems Man and Cybernetics Part C (Applications and Reviews)Fuzzy inference system learning by reinforcement methods
379 Citations1998Lionel Jouffe
Fuzzy Actor-Critic Learning (FACL) and Fuzzy Q-Learning are reinforcement learning methods based on dynamic programming (DP) principles and the genericity of these methods allows them to learn every kind of reinforcement learning problem.
Applied Soft ComputingOn learning algorithm selection for classification
374 Citations2006Shawkat Ali, Kate Smith‐Miles
This paper introduces a new method for learning algorithm evaluation and selection, with empirical results based on classification, to generate rules, using the rule-based learning algorithm C5.0, to describeWhich types of algorithms are suited to solving which types of classification problems.
European RadiologyDigital image processing
285 Citations1997Mathias Prokop, Cornelia Schaefer‐Prokop
Unsharp mask filtering (UMF) is a simple technique for improving image quality but suffers from serious drawbacks, such as the suppression of pathologic lesions or artifacts that may simulate pathology.
An Application of Reinforcement Learning to Dialogue Strategy Selection in a Spoken Dialogue System for Email
185 Citations2000Marilyn Walker Walker
Proceedings of 6th International Fuzzy Systems ConferenceFuzzy Q-learning
168 Citations2002P.-Y. Glorennec, Lionel Jouffe
This paper proposes an adaptation of Watkins' Q-learning for fuzzy inference systems where both the actions and the Q-functions are inferred from fuzzy rules, showing its effectiveness.
Fuzzy Q-learning and dynamical fuzzy Q-learning
90 Citations1994P.-Y. Glorennec
This paper proposes two reinforcement-based learning algorithms: 1) fuzzy Q- learning in an adaptation of Watkins' Q-learning for fuzzy inference systems; and 2) dynamical fuzzyQ-learning which eliminates some drawbacks of both Q-learnings.
Pattern RecognitionUnified formulation of a class of image thresholding techniques
84 Citations1996Hong Yan
It is shown that Otsu's image thresholding, Kittler and Illingworth's minimum errorresholding, and Huang and Wang's fuzzy thresholding methods can be derived under a similar mathematical formulation.
Journal of Artificial Intelligence ResearchAn Application of Reinforcement Learning to Dialogue Strategy Selection in a Spoken Dialogue System for Email
74 Citations2000M. A. Walker
A novel method by which a spoken dialogue system can learn to choose an optimal dialogue strategy from its experience interacting with human users, based on a combination of reinforcement learning and performance modeling of spoken dialogue systems.
Pattern Recognition LettersFuzzy divergence, probability measure of fuzzy events and image thresholding
68 Citations1992Dinabandhu Bhandari, Nikhil R. Pal +1 more
A tailored version of the probability measure of a fuzzy event is used for image segmentation and its application to clustering problems is indicated and applied to an object extraction problem.
Signal ProcessingMaximum entropy-based optimal threshold selection using deterministic reinforcement learning with controlled randomization
64 Citations2002Peng-Yeng Yin
It is shown that finding the optimal thresholds using the maximum entropy criterion is equivalent to learning an optimal policy of the RL problem and the powerful Q-learning algorithm, which is widely used in RL, can be employed to eradicate the computation burden of themaximum entropy-based thresholding methods.
Applied Soft ComputingObject detection in multi-modal images using genetic programming
57 Citations2004Bir Bhanu, Yingqiang Lin
It is shown that GP can synthesize effective composite operators consisting of pre-designed primitive operators and primitive features to effectively detect objects in images and the learned composite operators can be applied to the whole training image and other similar testing images.
Emotionally motivated reinforcement learning based controller
31 Citations2005Aladdin Ayesh
The main contribution of this paper is the use of fuzzy cognitive maps (FCM) to facilitate the modeling of emotions and inferencing for action selection in robotic control.
Using reinforcement learning for image thresholding
30 Citations2004Maryam Shokri, Hamid R. Tizhoosh
A reinforcement-learning concept to find the optimal threshold for digital images and can integrate human expert knowledge in an objective or subjective way to overcome the shortcomings of existing methods is introduced.
Application of reinforcement learning to balancing of Acrobot
18 Citations2003Junichiro Yoshimoto, Shin Ishii +1 more
This article discusses applying the RL to the task of balancing control of the Acrobot, and shows that the method is able to achieve fairly good control with a small number of trials.
Applied Soft ComputingFast adapting value estimation-based hybrid architecture for searching the world-wide web
18 Citations2002István Kókai, András Lörincz
A hybrid architecture that combines soft support vector classification and reinforcement learning for value estimation is introduced for the evaluation of a link and its neighboring links, called the context of a document.
Q(Λ)-based image thresholding
16 Citations2004Maryam Shokri, Hamid R. Tizhoosh
This paper introduces a new method for image thresholding using reinforcement learning as an effective way to find the optimal threshold and Q(Λ) is implemented as a learning algorithm to achieve more accurate results.
Canadian Conference on Computer and Robot VisionQ(lambda)-Based Image Thresholding.
10 Citations2004Maryam Shokri, Hamid R. Tizhoosh
It was clearly showed that the ALBI grade was associated with toxicity and survival after this treatment, and the baseline ALBI score and variations of ALBI are prognostic after SIRT.
Analog And Digital Communication
7 Citations2023Bharathy G.T., S. Bhavanisankari +2 more
Advanced information and knowledge processingReinforcement Agents for E-Learning Applications
3 Citations2007Hamid R. Tizhoosh, Maryam Shokri +1 more
The ability of learning from interaction with a dynamic environment and using reward and punishment independent of any training data set makes reinforcement learning a suitable tool for e-learning situations, where subjective user feedback can easily be translated into a reinforcement signal.