Chi-Kin Leung

PERFORMANCE ANALYSIS FOR A CLASS OF ITERATIVE IMAGE THRESHOLDING ALGORITHMS

A performance analysis procedure that analyses the properties of a class of iterative image thresholding algorithms is described. The image under consideration is modeled as consisting of two maximum-entropy primary images, each of which has a quasi-Gaussian probability density function. Three iterative thresholding algorithms identified to share a common iteration architecture are employed for thresholding 4595 synthetic images and 24 practical images. The average performance characteristics including accuracy, stability, speed and consistency are analysed and compared among the algorithms. Both analysis and practical thresholding results are presented.

Maximum segmented image information thresholding

Abstract Utilizing information theory and considering image segmentation from a communication perspective, the image segmentation process is interpreted as a data processing step that operates on a gray-scale image and produces a segmented image. It is shown that the segmented image contains a certain amount of information about the scene, which is defined assegmented image information(SII). It is proposed that the SII should be maximized when an image is thresholded, and this is known as themaximum segmented image information(MSII) thresholding criterion. The MSII thresholding criterion possesses better properties as compared with theminimum error(MINE) and theuniform error(UNFE) thresholding criteria. Based on the MSII thresholding criterion, an MSII thresholding algorithm is proposed for the thresholding of real images. The MSII thresholding algorithm is evaluated against several well-known thresholding algorithms. The good thresholding results of both synthetic and real images confirm the capabilities of the proposed MSII thresholding algorithm.

Image segmentation by edge pixel classification with maximum entropy

Image segmentation is a process to classify image pixels into different classes according to some pre-defined criterion. An entropy based image segmentation method is proposed to segment a gray-scale image. The method starts with an arbitrary template. An index called Gray-scale Image Entropy (GIE) is employed to measure the degree of resemblance between the template and the true scene that gives rise to the gray-scale image. The classification status of the edge pixels in the template is modified in such a way as to maximize the GIE. By repeatedly processing all the edge pixels until a termination condition is met, the template would be changed to a configuration that closely resembles the true scene. This optimum template (in an entropy sense) is taken to be the desired segmented image. Investigation results from simulation study and the segmentation of practical images demonstrate the feasibility of the proposed method.

Maximum segmented-scene spatial entropy thresholding

The segmented-scene spatial entropy (SSE) is defined as the amount of information contained in the spatial structure of a segmented scene resulting from segmenting an image. An automatic, nonparametric, unsupervised thresholding algorithm that maximizes the SSE of an image is described, and this algorithm is known as the maximum segmented-scene spatial entropy (MSSE) thresholding algorithm. It is shown that the MSSE-thresholded image contains the maximum amount of information about the original scene and hence good thresholding results are warranted. Simulation and practical results are presented to illustrate the improvement in performance as compared to some other histogram-based thresholding algorithms.

Image segmentation using maximum entropy method

Segmentation of a composite image which contains two simple subimages is described. The a-priori knowledge about the two simple subimages is that they possess the maximum amount of entropy. The probability density functions (pdfs) of these image pixels are shown to be of the quasi-Gaussian form. Parameters for the pdf are estimated and then the maximum likelihood ratio test is applied to segmentation. An iterative algorithm is employed to improve the segmentation accuracy. Extension of this method to the segmentation of images with arbitrary pdfs is discussed.<<ETX>>

Ultrasonic detection using wideband discrete wavelet transform

This paper describes the design of a wideband spatial processor for the detection of a straight-line object by an ultrasonic pulse-echo detection system. An ultrasonic pulse is transmitted from the transducer and the two wavelets diffracted from the two end points of the straight-line object are received by three spatially separated receivers. Three stages of signal processing are carried out. At the first stage, the mother wavelet operator generates three sets of two-dimensional wavelet coefficients. At the second stage, cross wavelet transforms are performed on wavelet coefficients obtained in the first stage. At the third stage, cross wavelet transforms are performed on cross wavelet coefficients obtained in the second stage. As a result of this three-stage operation, a high-resolution image of the environment is generated and the range and bearing of the two end points of the straight-line object are obtained. A simulation program is developed to investigate the processing algorithm in an ultrasonic detection environment.

An iterative image segmentation algorithm utilizing spatial information

An iterative image segmentation algorithm that segments an image on a pixel-by-pixel basis is described. The observation information to be utilized is the joint gray level values of the pixel to be segmented and those of its neighborhood pixels. The iterative process is initialized by thresholding the image with Otsus (1979) method. Each pixel is segmented into a class when the a posteriori probability, conditioned on the observation information, that it belongs to this class is a maximum. The newly segmented image is employed to re-estimate the a posteriori probabilities and the segmentation process is repeated until there is no further pixel classification change in a particular run. Among those segmented images generated in the iterative process, the best segmented image is chosen, according to a maximum entropy criterion. Simulation studies demonstrate that the proposed algorithm can achieve very significant improvement in segmentation performance as compared to the more popular thresholds approach. Furthermore, the performance is neither sensitive to the initial threshold value nor the form of the probability density function of the image. Segmentation of practical images also demonstrates that the proposed algorithm is capable of good segmentation results for real-life images.

On prioritizing the delivery of short videos clips

In interactive video-on-demand (IVOD) applications, service availability and response times are more visible to the user than the underlying data throughput. In this paper, a resource model for incoming requests is proposed. Based on the resource model, the performance of several admission policies for short video requests is evaluated in terms of probability of admission, queuing delay, and throughput. Simulation results show that prioritizing requests can increase the number of admitted requests and reduce the queuing time. The technique proposed in the paper is independent of the underlying disk scheduling technique, hence it can be employed to improve the user-perceived performance of VOD servers under various system loading conditions.

Locating bus cable break point in a local area network

Abstract A measuring system which can accurately locate an inaccessible break point over a local area network bus cable is described. The system utilizes sinusoidal signals instead of conventional narrow pulses as input test signals. It is shown that the variation of the amplitude with frequency in the resultant sinusoidal wave established over the cable length is periodic and that the periodicity is related to the break point distance. Thus the distance to the break point can be determined by measuring this period. Practical tests on a number of cable segments confirmed the viability of the approach. On account of its accuracy and modest equipment requirement, the method is valuable for break point location measurement over an Ethernet environment.

Image segmentation with scalable spatial information

A general approach is proposed for the design of image segmentation algorithms utilizing spatial information which is the combined properties of a collection of neighborhood pixels. With different types of properties and different number of neighborhood pixels being utilized, segmentation algorithms with different speed and accuracy performance can be designed. Six algorithms have been implemented with their performance investigated and compared.