Turgay Celik

Bio: Turgay Celik is an academic researcher from University of the Witwatersrand. The author has contributed to research in topics: Change detection & Pixel. The author has an hindex of 33, co-authored 118 publications receiving 4248 citations. Previous affiliations of Turgay Celik include Nanyang Technological University & University of Warwick.

Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and $k$ -Means Clustering

Abstract: In this letter, we propose a novel technique for unsupervised change detection in multitemporal satellite images using principal component analysis (PCA) and k-means clustering. The difference image is partitioned into h times h nonoverlapping blocks. S, S les h2, orthonormal eigenvectors are extracted through PCA of h times h nonoverlapping block set to create an eigenvector space. Each pixel in the difference image is represented with an S-dimensional feature vector which is the projection of h times h difference image data onto the generated eigenvector space. The change detection is achieved by partitioning the feature vector space into two clusters using k-means clustering with k = 2 and then assigning each pixel to the one of the two clusters by using the minimum Euclidean distance between the pixel's feature vector and mean feature vector of clusters. Experimental results confirm the effectiveness of the proposed approach.

Fire detection in video sequences using a generic color model

Abstract: In this paper, a rule-based generic color model for flame pixel classification is proposed. The proposed algorithm uses YCbCr color space to separate the luminance from the chrominance more effectively than color spaces such as RGB or rgb. The performance of the proposed algorithm is tested on two sets of images, one of which contains fire, the other containing fire-like regions. The proposed method achieves up to 99% fire detection rate. The results are compared with two other methods in the literature and the proposed method is shown to have both a higher detection rate and a lower false alarm rate. Furthermore the proposed color model can be used for real-time fire detection in color video sequences, and we also present results for segmentation of fire in video using only the color model proposed in this paper.

Contextual and Variational Contrast Enhancement

Abstract: This paper proposes an algorithm that enhances the contrast of an input image using interpixel contextual information. The algorithm uses a 2-D histogram of the input image constructed using a mutual relationship between each pixel and its neighboring pixels. A smooth 2-D target histogram is obtained by minimizing the sum of Frobenius norms of the differences from the input histogram and the uniformly distributed histogram. The enhancement is achieved by mapping the diagonal elements of the input histogram to the diagonal elements of the target histogram. Experimental results show that the algorithm produces better or comparable enhanced images than four state-of-the-art algorithms.

Fire detection using statistical color model in video sequences

Abstract: In this paper, we propose a real-time fire-detector that combines foreground object information with color pixel statistics of fire. Simple adaptive background model of the scene is generated by using three Gaussian distributions, where each distribution corresponds to the pixel statistics in the respective color channel. The foreground information is extracted by using adaptive background subtraction algorithm, and then verified by the statistical fire color model to determine whether the detected foreground object is a fire candidate or not. A generic fire color model is constructed by statistical analysis of the sample images containing fire pixels. The first contribution of the paper is the application of real-time adaptive background subtraction method that aids the segmentation of the fire candidate pixels from the background. The second contribution is the use of a generic statistical model for refined fire-pixel classification. The two processes are combined to form the fire detection system and applied for the detection of fire in the consecutive frames of video sequences. The frame-processing rate of the detector is about 40 fps with image size of 176x144 pixels, and the algorithm's correct detection rate is 98.89%.

Automatic Image Equalization and Contrast Enhancement Using Gaussian Mixture Modeling

Abstract: In this paper, we propose an adaptive image equalization algorithm that automatically enhances the contrast in an input image. The algorithm uses the Gaussian mixture model to model the image gray-level distribution, and the intersection points of the Gaussian components in the model are used to partition the dynamic range of the image into input gray-level intervals. The contrast equalized image is generated by transforming the pixels' gray levels in each input interval to the appropriate output gray-level interval according to the dominant Gaussian component and the cumulative distribution function of the input interval. To take account of the hypothesis that homogeneous regions in the image represent homogeneous silences (or set of Gaussian components) in the image histogram, the Gaussian components with small variances are weighted with smaller values than the Gaussian components with larger variances, and the gray-level distribution is also used to weight the components in the mapping of the input interval to the output interval. Experimental results show that the proposed algorithm produces better or comparable enhanced images than several state-of-the-art algorithms. Unlike the other algorithms, the proposed algorithm is free of parameter setting for a given dynamic range of the enhanced image and can be applied to a wide range of image types.

Machine learning

Abstract: Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Pattern Recognition and Machine Learning

Abstract: Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Computer and Robot Vision

Abstract: Computer and Robot Vision Vol. 1, by R.M. Haralick and Linda G. Shapiro, Addison-Wesley, 1992, ISBN 0-201-10887-1.

LIME: Low-Light Image Enhancement via Illumination Map Estimation

Abstract: When one captures images in low-light conditions, the images often suffer from low visibility. Besides degrading the visual aesthetics of images, this poor quality may also significantly degenerate the performance of many computer vision and multimedia algorithms that are primarily designed for high-quality inputs. In this paper, we propose a simple yet effective low-light image enhancement (LIME) method. More concretely, the illumination of each pixel is first estimated individually by finding the maximum value in R, G, and B channels. Furthermore, we refine the initial illumination map by imposing a structure prior on it, as the final illumination map. Having the well-constructed illumination map, the enhancement can be achieved accordingly. Experiments on a number of challenging low-light images are present to reveal the efficacy of our LIME and show its superiority over several state-of-the-arts in terms of enhancement quality and efficiency.

TFOS DEWS II Diagnostic Methodology report

Abstract: The role of the Tear Film and Ocular Surface Society (TFOS) Dry Eye Workshop (DEWS) II Diagnostic Methodology Subcommittee was 1) to identify tests used to diagnose and monitor dry eye disease (DED), 2) to identify those most appropriate to fulfil the definition of DED and its sub-classifications, 3) to propose the most appropriate order and technique to conduct these tests in a clinical setting, and 4) to provide a differential diagnosis for DED and distinguish conditions where DED is a comorbidity. Prior to diagnosis, it is important to exclude conditions that can mimic DED with the aid of triaging questions. Symptom screening with the DEQ-5 or OSDI confirms that a patient might have DED and triggers the conduct of diagnostic tests of (ideally non-invasive) breakup time, osmolarity and ocular surface staining with fluorescein and lissamine green (observing the cornea, conjunctiva and eyelid margin). Meibomian gland dysfunction, lipid thickness/dynamics and tear volume assessment and their severity allow sub-classification of DED (as predominantly evaporative or aqueous deficient) which informs the management of DED. Videos of these diagnostic and sub-classification techniques are available on the TFOS website. It is envisaged that the identification of the key tests to diagnose and monitor DED and its sub-classifications will inform future epidemiological studies and management clinical trials, improving comparability, and enabling identification of the sub-classification of DED in which different management strategies are most efficacious.