Cliff Wang

Supervised Multispectral Image Segmentation using Active Contours

Active contours have been widely used as image segmentation methods. The use of level set theory has provided more flexibility and convenience for the implementation of active contours. However, traditional active contour models have some limitations on the segmentation of complicated images whose sub-regions consist of multiple components. The segmentation of multispectral images is even a more difficult problem. We propose an advanced active contour model using the statistics of image intensity based on a multivariate mixture density model. The proposed active contour model shows a robust segmentation capability on the images that traditional segmentation methods cannot properly partition. Numerical experiments with synthetic and real images are presented.

Experiments and Results

This chapter describes the experimental results of the MRF-based method for homology detection and fold recognition including alignment accuracy, success rate, running time and contribution of some important features. This chapter also compares the MRF-based method with currently popular PSSM- and HMM-based methods such as HHpred, HHblits and FFAS, in terms of alignment accuracy and success rate of homology detection and fold recognition.

Edge detection in gated cardiac nuclear medicine images

Mean field annealing using a piecewise linear model was applied to gated cardiac nuclear medicine images as a preprocessing tool for image smoothing and noise reduction. A second derivative operator was then used to extract the edges for ventricle boundary estimation. Combined with the user input initial boundary estimate, the extracted edge information was used to find a minimum cost boundary, which was optimum with regards to boundary smoothness and the boundary edge strength.<<ETX>>

Secure Tracking in Sensor Networks

Target tracking is a canonical issue in sensor networks research. However, tracking security has gained little or no attention. Once a sensor node is compromised, it will be able to inject false location information into the network, and those nodes receiving such information will suffer greatly in terms of tracking precision. This paper, to the best of our knowledge, is the first to explore the topic of security in the context of Bayesian tracking for sensor networks. We propose to activate more than one nodes at each time step, and use a relaxation labeling algorithm to detect malicious nodes whose reports are then removed. Simulations based on both linear and nonlinear motion models demonstrate that out algorithm works better than simply averaging over the results based on the redundant sets of nodes.

A New Relaxation Labeling Architecture for Secure Localization in Sensor Networks

In this paper, a new strategy is proposed to defend against colluding malicious nodes in a sensor network. The new strategy is based on a new relaxation labeling algorithm to classify nodes into benign or malicious ones. Only reports from benign nodes can then be used to perform localization and obtain accurate results. Experimental results based on simulations and field experiments illustrate the performance of the algorithm.

Human-Computer Interaction

The main objective of this chapter is to summarize the authors’ recent work which studied the human-computer interaction for sketch-based passwords. However, before discussing the previous results, the features used specifically for implementation of SKS are discussed, and an overview of the database used for the HCI study (and some experiments and results in the subsequent chapter) is provided. The features used for the application of sketch-based passwords incorporates four fundamental properties: (1) shape, (2) direction, (3) order, and (4) pressure, which comprise a local 5D feature descriptor at every sketch sample point. The database used for this HCI study was constructed by the authors in (Riggan et al., A human factors study of graphical passwords using biometrics. Proc. of the 36th German Conf. on Pattern Recognition, 2014). This database considers a sufficiently large and complete set of users and many variations of sketch-based passwords. The previous HCI is useful to understand the connection between human perception and certain similarity measures, specifically for sketch-based authentication.

DICOM 3.0 image display and processing tool for teleradiology

A typical teleradiology system consists of four sub-systems: (1) image acquisition, (2) image transmission, (3) image viewing, and (4) teleconferencing. An image viewing and processing tool is a very important part of the system. A successful teleradiology system would require an effective image display and processing tool based on a standard user friendly GUI. DICOM 3.0 defines vender independent data formats and data transfers for digital medical images. It has been widely supported by industry since the first draft of DICOM 3.0 standard. In this paper, we present a DICOM 3.0 image display and processing tool for teleradiology and teleconsulting. The system provides the user a flexible image display format and a powerful set of image processing tools. A DICOM panel displays the grouped information of patient, study, result, and acquisition setting. This display and processing tool is designed for both clinical and research use.

Sketch-Based Authentication

This chapter considers two different recognition/matching algorithms, namely dynamic time warping (DTW) and Simple K-Space (SKS). Both algorithms utilize classical pattern recognition techniques to the novel application of sketch-based password. While more sophisticated techniques are possible, DTW and SKS are chosen because the inherent nature of the underlying problem is easily perceived using the approaches. Plus, the fundamental concepts discussed in the chapter generalize beyond the specific algorithmic implementation. The overall sketch-based authentication framework is outlined in this chapter using DTW and SKS. The respective implementation details are rigorously discussed, from which the philosophies are compared and contrasted. In particular, these methods have a similar objective, but different parameterizations, operating spaces, and computational complexities.

Efficiency, Uniqueness, and Robustness

There are (at least) three primary considerations when designing a sketch-based authentication system: computational complexity, security, and tolerance. In this chapter, using the generalized Simple K-Space (SKS) approach, the efficiency of model construction is studied. There are two algorithmic implementations to construct this model: the direct/forward approach (or primal algorithm) or the indirect/backward (or dual algorithm). The efficiency of constructing the SKS model is important because the model is most definitely the computational bottleneck of this method. Here, the dual algorithm is shown to enable the use of a model approximation that significantly improves the efficiency for computing the high dimensional model representation. Also, the tradeoff between security and usability is analyzed using SKS. In particular, it is demonstrated that as smoothing parameters of the SKS model asymptotically approach zero, the model is unique and provides “perfect security.” However, we also demonstrate the model is generally not unique, but more robust, when a sufficient amount of smoothing occurs over the model. Therefore, SKS is shown to explicitly provide fine-tuned controllability of the security/usability tradeoff.

Fundamentals of Sketch-Based Passwords

Keeping data, which is stored on a variety of information systems, e.g. networks, desktop and laptop computers, personal digital assistants (PDAs), smartphones, tablets, or any other electronic media, private is essential for individual security, corporate security, and national security. While a majority of authentication mechanisms remain the same as they were more than a decade ago, attacks on information system infrastructures are becoming more sophisticated. Many, including financial institution, government agencies, and large organization, have always invested both time and money into improved security solutions. One emerging topic, namely sketch-based passwords, is the subject of this brief. In this chapter, the need for alternative security solutions is greatly emphasized, and the fundamental concept of using sketch-based passwords (one possible solution) is introduced. Additionally, the major contributions and results presented in this brief are summarized.