Hong Man

• Stevens Institute of Technology

Sentiment Quantification aims to detect the overall sentiment polarity of users from a set of reviews corresponding to a target. Existing methods equally treat and aggregate individual reviews' sentiment to judge the overall sentiment polarity. However, the confidence of each review is not equal in sentiment quantification where sentiment perturbat...

The task of cross-modal image retrieval has recently attracted considerable research attention. In real-world scenarios, keyword-based queries issued by users are usually short and have broad semantics. Therefore, semantic diversity is as important as retrieval accuracy in such user-oriented services, which improves user experience. However, most t...

Background objects and textures in real-world video sequences often pose great challenges for human action and facial expression recognition. This paper proposes a mixture statistic metric learning for recognizing human actions and facial expressions in realistic “in the wild” scenarios. In the proposed method, multiple statistics, including tempor...

Opportunistic cognitive routing coupled with dual-stage collaborative spectrum sensing in multi-channel multi-hop cognitive radio ad hoc networks (CRAHNs) is investigated in this work. Due to the dynamics of available spectrum and complex radio environment in CRAHNs, an opportunistic cognitive routing protocol is proposed, that exploits a dual-stag...

Feature selection aims to gain relevant features for improved classification performance and remove redundant features for reduced computational cost. How to balance these two factors is a problem especially when the categorical labels are costly to obtain. In this paper, we address this problem using semisupervised learning method and propose a ma...

This paper proposes a statistical adaptive metric learning method by exploring various selections and combinations of multiple statistics in a unified metric learning framework. Most statistics have certain advantages in specific controlled environments, and systematic selections and combinations can adapt them to more realistic “in the wild” scena...

Accurate and accelerated MRI tissue recognition is a crucial preprocessing for real-time 3d tissue modeling and medical diagnosis. This paper proposed an information de-correlated clustering algorithm implemented by variational level set method for fast tissue segmentation. The key idea is to design a local correlation term between original image a...

This paper considers Multi-Arms Restless Bandits problem, where each arm have time varying rewards generated from unknown two-states discrete time Markov process. Each chain is assumed irreducible, aperiodic, and non-reactive to agent actions. Optimal solution or constant value approximation to all instances of Restless Bandits problem does not exi...

Great variances in visual features often present significant challenges in human action recognitions. To address this common problem, this paper proposes a statistical adaptive metric learning (SAML) method by exploring various selections and combinations of multiple statistics in a unified metric learning framework. Most statistics have certain ad...

A new front-end feature extraction scheme creating so called LDA-projected magnitude spectrum (L-PMS) features is proposed for speaker recognition systems. Mainstream feature extraction schemes usually use filter-bank or linear predictive coding (LPC) in the process of converting high-dimensional speech data into low-dimensional feature vectors, wh...

Cognitive routing coupled with cooperative spectrum sensing in multi-hop Cognitive Radio Ad Hoc Network (CRAHN) is investigated. Recognizing the spectrum dynamics and the problems of hidden terminal and shadow fading in CRAHNs, we propose an opportunistic routing protocol that exploits a Dual-stage Collaborative Spectrum Sensing (DCSS) scheme to im...

Sparse representation, which uses dictionary atoms to reconstruct input vectors, has been studied intensively in recent years. A proper dictionary is a key for the success of sparse representation. In this paper, an active dictionary learning (ADL) method is introduced, in which classification error and reconstruction error are considered as the ac...

Sparse-representation-based classification (SRC), which classifies data based on the sparse reconstruction error, has been a new technique in pattern recognition. However, the computation cost for sparse coding is heavy in real applications. In this paper, various dimension reduction methods are studied in the context of SRC to improve classificati...

In this paper, a hybrid learning model of imbalanced evolving self-organizing maps (IESOMs) is proposed to address the imbalanced learning problems. In our approach, we propose to modify the classic SOM learning rule to search the winner neuron based on energy function by minimally reducing local error in the competitive learning phase. The advanta...

Because of the lack of disciplined and efficient mechanisms, most modern area charge-coupled device-based barcode scanning technologies are not capable of handling out-of-focus (OOF) image blur and rely heavily on camera systems for capturing good quality, well-focused barcode images. In this paper, we present a novel linear barcode scanning system...

Despite the increasing importance of robotics, there is a significant challenge involved in teaching this to undergraduate students in biomedical engineering (BME) and other related disciplines in which robotics techniques could be readily applied. This paper addresses this challenge through the development and pilot testing of a bio-microrobotics...

Fifth generation wireless systems (5G) must achieve high user Quality of Experience (QoE) in order to compete for market share. Each candidate 5G wireless radio frequency (RF) band offers advantages such as longer range or higher data rate than 2G, 3G, and 4G, but no single band or air interface standard by itself fully achieves ubiquitous levels o...

In this paper, a novel statistical tool, stochastic context-free models (SCFMs), is introduced to model and analyze brain voxel activation in fMRI time series. SCFMs characterize the dynamic process where Blood-oxygen-level dependent (BOLD) responses are assumed to be driven by brain voxel activation in pre-designed experiments. Classical state spa...

Spectrum Sensing is an intensively studied topic in cognitive radio to locate unoccupied spectrum for improved channel utilization. However, the problem becomes more challenging in wideband spectrum sensing due to the limitation of hardware operational bandwidth. In this paper, we introduce a cooperative compressive spectrum sensing scheme to monit...

Automatic modulation classification (AMC) is an important component in cognitive radio and many efforts have been made to improve the AMC's successful classification rate, especially when the environment is noisy. The cyclic feature has excellent resiliency to noise, so it has been frequently adopted as the feature for AMC. In this paper, in order...

Wireless capsule endoscopy (WCE) is a newly booming technology on gastrointestinal (GI) examinations. After the patient swallows the capsule, it starts taking video of the entire interior of the digestive tract, from the esophagus, stomach to small intestine, and colon. It transmits sequences of color images back to the computer, and hence allows t...

In this paper, we propose an iterative self-organizing map (SOM) approach with robust distance estimation (ISOMRD) for spatial outlier detection. Generally speaking, spatial outliers are irregular data instances which have significantly distinct non-spatial attribute values compared to their spatial neighbors. In our proposed approach, we adopt SOM...

Recognizing human-object interactions in videos is a very challenging problem in computer vision research. There are two major difficulties lying in this task: (1) The detection of human body parts and objects is usually affected by the quality of the videos, for instance, low resolutions of the videos, camera motions, and blurring frames caused by...

Cyclic spectrum feature is one of the most popular features used in automatic modulation recognition (AMR) due to its excellent resiliency to noise. However, extracting cyclic features from wireless signals always requires at least Nyquist rate in traditional. What's more, to better capture cyclostationarity for modulation classification, the sampl...

To recognize concurrent human activities in videos, we proposed a cognitive semantics based novel event representation for small human group detection and event recognition. Given a video with human detection and tracking results, the video is firstly described by cognitive linguistic primitives, including "paths", "places", "things", "actions", an...

In this paper, we propose a spatio-temporal dependencies learning (STDL) method for action recognition. Inspired by self-organizing map, our method can learn implicit spatial-temporal dependencies from sequential action feature sets while preserving the intrinsic topologies characterized in human actions. A further advantage is its ability to proje...

In machine learning and pattern recognition, feature selection has been a very active topic in the literature. Unsupervised feature selection is challenging due to the lack of label which would supply the categorical information. How to define an appropriate metric is the key for feature selection. In this paper, we propose a "filter" method for un...

In this paper, a Dynamic Structure Preserving Map (DSPM) is proposed to effectively recognize human actions in video sequences. Inspired by the latest feature learning methods, we modified and improved the adaptive learning procedure in self-organizing map (SOM) to capture dynamics of best matching neurons through Markov random walk. The DSPM can l...

Electro-Optic (EO) image sensors exhibit the properties of high resolution and low noise level, but they cannot reflect information about the temperature of objects and do not work in dark environments. On the other hand, infrared (IR) image sensors exhibit the properties of low resolution and high noise level, but IR images can reflect information...

As a uncommon biometric modality, human gait recognition has a great advantage of identify people at a distance without high resolution images. It has attracted much attention in recent years, especially in the fields of computer vision and remote sensing. In this paper, we propose a human gait recognition framework that consists of a reliable back...

Electro-optic (EO) image sensors exhibit the properties of high resolution and low noise level at daytime, but they do not work in dark environments. Infrared (IR) image sensors exhibit poor resolution and cannot separate objects with similar temperature. Therefore, we propose a novel framework of IR image enhancement based on the information (e.g....

Small group people activity recognition has attracted much attention in computer vision community in recent years, since it has great potential in public security applications. Comparing to single human activity recognition, group human activity recognition has much more challenges, such as mutual occlusions between different people, the varying gr...

In this paper, we propose a novel method SOMKE, for kernel density estimation (KDE) over data streams based on sequences of self-organizing map (SOM). In many stream data mining applications, the traditional KDE methods are infeasible because of the high computational cost, processing time, and memory requirement. To reduce the time and space compl...

Co-training is a well-known semi-supervised learning technique that applies two basic learners to train the data source, which uses the most confident unlabeled data to augment labeled data in the learning process. In the paper, we use the diversity of class probability estimation (DCPE) between two learners and propose the DCPE co-training approac...

Feature selection, which aims to obtain valuable feature subsets, has been an active topic for years. How to design an evaluating metric is the key for feature selection. In this paper, we address this problem using imputation quality to search for the meaningful features and propose feature selection via sparse imputation (FSSI) method. The key id...

To enhance network security, we propose a secret key generation and distribution method for multihop wireless OFDM networks. In this method, the inherent physical features of wireless channel signatures, including randomness and reciprocity, are exploited to generate secret keys. In addition, a network coding approach is introduced for key distribu...

Spectrum sensing is one of the most challenging problems in cognitive radio systems. It is frequently impractical to implement theoretical methods due to the limitation of the existing hardware operational bandwidth. To solve this problem, an emerging technique, compressive sensing (CS), is introduced to cognitive radio field so that only compressi...

Electro-optic (EO) images exhibit the properties of high resolution and low noise level, while it is a challenge to distinguish objects with infrared (IR), especially for objects with similar temperatures. In earlier work, we proposed a novel framework for IR image enhancement based on the information (e.g., edge) from EO images. Our framework supe...

Electro-optic (EO) images exhibit the properties of high resolution and low noise level, while it is a challenge to distinguish objects at night through infrared (IR) images, especially for objects with a similar temperature. Therefore, we will propose a novel framework of IR image enhancement based on the information (e.g., edge) from EO images, w...

Supported by an NSF CCLI award, we have developed teaching materials based on a case study on a pill-sized robot in gastro-intestinal (GI) tract to teach undergraduate microrobotics and also principles of robot programming and navigation. The case study consists of a lecture unit and a laboratory module. The lecture unit introduces commercial capsu...

This paper introduces two unsupervised learning methods for analyzing functional magnetic resonance imaging (fMRI) data based on hidden Markov model (HMM). HMM approach is focused on capturing the first-order statistical evolution among the samples of a voxel time series, and it can provide a complimentary perspective of the BOLD signals. Two-state...

In this paper, we introduce a learning based cognitive radio receiver to automatically demodulate several types of modulated signals without sophisticated digital signal pre-processing. Our embedded learning engine can automatically learn the signal features and then achieve signal demodulation through feature-based classification. The proposed dem...

Multiple modalities sensor fusion has been widely employed in various surveillance and military applications. A variety of image fusion techniques including PCA, wavelet, curvelet and HSV has been proposed in recent years to improve human visual perception for object detection. One of the main challenges for visible and infrared image fusion is to...

Sparse coding, which produces a vector representation based on sparse linear combination of dictionary atoms, has been widely applied in signal processing, data mining and neuroscience. Constructing a proper dictionary for sparse coding is a common challenging problem. In this paper, we treat dictionary learning as an unsupervised learning process,...

Multiple Self-Organizing Maps (MSOMs) based classification methods are able to combine the advantages of both unsupervised and supervised learning mechanisms. Specifically, unsupervised SOM can search for similar properties from input data space and generate data clusters within each class, while supervised SOM can be trained from the data via labe...

Automatic modulation recognition (AMR) and demodulation are two essential components in cognitive radio receivers. This paper proposes a novel method based on MSOM neural networks to automatically recognize the modulation type and demodulate the radio signal at the same time. This efficient method is directly applied to the normalized radio signal...

In this paper, we present a method of utilizing channel diversity to increase secrecy capacity in wireless communication. With the presence of channel diversity, an intended receiver can achieve a relatively high secrecy capacity even at low SNRs. We present a theoretical analysis on the outage probability at a normalized target secrecy capacity in...

The fusion of images captured from multi-modality sensors has been studied for many years. It is aiming at combining multiple sources together to maximize the meaningful information and reduce the redundancy. Meanwhile, sparse representation of images has been attracting more and more attentions. It has been effectively utilized on image reconstruc...

Ultrasound has shown a great potential for serving as an outcome measure for localized scleroderma. High spatial resolution of ultrasound will aid to evaluate Echogenicity, Vascularity index and tissue thickness of localized scleroderma. Echogenicity is higher when the ultrasound waves reflected are lesser in value from any particular body organ. V...

Organisations and individuals benefit when wireless networks are protected. After assessing the risks associated with wireless technologies, organisations can reduce the risks by applying countermeasures to address specific threats and vulnerabilities. These countermeasures include management, operational and technical controls. While these counter...

Nanotechnology, the ability to leverage and exploit fundamental processes at the nanometer length scale, suggests the potential for a technological revolution. To sustain and propagate technologies at the nanoscale, continued efforts toward understanding the fundamental principles governing nano-science must be coupled with a focus on nano-engineer...

We present instructional materials to teach bio-medical engineering students about the design and control of a capsule robot operating in the human's GI tract. A design example to conceptually build such a micro-robot is first presented, and a laboratory module is then developed to demonstrate robot navigation techniques. Medical considerations suc...

The broadcast nature of the wireless medium poses a serious challenge to the security of wireless communications and networks. Traditional computational security strategies always fail to prevent the eavesdropper from overhearing the communications. Recently, the unique characteristics of wireless channel fingerprints, e.g. randomness and reciproci...

Secure wireless communication is a challenging problem due to the shared nature of the wireless medium and the dynamic channel. Most of the existing security mechanisms focus on traditional cryptographic schemes. In recent years, features of multi-path channels, such as randomness, coherence and reciprocity, have driven researchers to exploit their...

Localized Pediatric Scleroderma is an auto-immune disease, the most common type of scleroderma in children and affects the skin, only occasionally spreading to the underlying muscles. Scleroderma has prevalence rate is 0.11% and occurs approximately 1 in every 906 persons in the United States of America. Ultrasound imaging, a non-invasive and less...

Human motion change detection is a challenging task for a surveillance sensor system. Major challenges include complex scenes with a large amount of targets and confusors, and complex motion behaviors of different human objects. Human motion change detection and understanding have been intensively studied over the past decades. In this paper, we pr...

Automatic modulation recognition (AMR) of com-munication signals is a critical and challenging task in cogni-tive radio systems. In this work, classifications of four digital modulation types, including BPSK, QPSK, GMSK and 2FSK, are investigated. From the received radio signal, a set of cyclic spectrum features are first calculated, and a principa...

The fusion of images captured from Electrical-Optical (EO) and Infra-Red (IR) cameras has been extensively studied for military applications in recent years. In this paper, we propose a novel wavelet-based framework for online fusion of EO and IR image sequences. The proposed framework provides multiple fusion rules for image fusion as well as a no...

In this paper, a new method is proposed for removing and restoring random-valued impulse noise in images. This approach is based on a similar neighbor criterion, in which any pixel to be considered as an original pixel it should have sufficient numbers of similar neighboring pixels in a set of filtering windows. Compared with other well known metho...

The imbalanced learning problem (learning from imbalanced data) presents a significant new challenge to the pattern recognition and machine learning society because in most instances real-world data is imbalanced. When considering military applications, the imbalanced learning problem becomes much more critical because such skewed distributions nor...

Systems engineering approaches are employed to measure and to analyze vulnerabilities of military tactical RF wireless networks. The goal is to develop smart and innovative performance benchmarks through electronic warfare (EW) modeling and simulation scenarios. Systematic systems engineering approaches with radio frequency (RF) electronic warfare...

Co-training is a semi-supervised learning technique used to recover the unlabeled data based on two base learners. The normal co-training approaches use the most confidently recovered unlabeled data to augment the training data. In this paper, we investigate the co-training approaches with a focus on the diversity issue and propose the diversity of...

In this paper, we propose an iterative self-organizing map approach for spatial outlier detection (IterativeSOMSO). IterativeSOMSO method can address high dimensional problems for spatial attributes and accurately detect spatial outliers with irregular features. Detection of spatial outliers facilitates further discovery of spatial distribution and...

Anomaly detection in peer-to-peer (P2P) networks is generally difficult due to the large number of users in the network. Exhaustive probing on each user is extremely unrealistic. Besides, unlike hierarchical systems, the infrastructure of a P2P network is flat, which makes multi-casting based probing schemes impossible. Most P2P security research f...

A cascade of filtering windows is implemented iteratively for removing random-valued impulse noise in heavily corrupted images. This method is based on the peer group concept (PGC), so a pixel is considered as noise-free if and only if for each window size, there exists a peer group of certain threshold cardinality for it. Otherwise, the pixel is c...

We present the progress of our NSF CCLI project to design teaching materials on micro/nanorobotics for biomedical engineering students. We have developed a case study and a laboratory module, both of which are centered on a vitamin pill sized microrobot navigating in the human's GI tract. In particular, we built a simulation module in Webots 3D sim...

Social network analysis (SNA), originally introduced to provide a mathematical framework for analyzing human interactions and economic relationships, has recently been successfully applied to characterizing information propagation in wireless networks. In this paper, we introduce a SNA method as a new approach to build an intrusion detection system...

Detection and tracking of a varying number of people is very essential in surveillance sensor systems. In the real applications, due to various human appearance and confessors, as well as various environment conditions, multiple targets detection and tracking become even more challenging. During this year, our major contributions of multiple target...

This paper characterizes key semantics issues in shaping cognitive radio behavior to realize social contracts and business logic via cognitive radio behavioral semantics, with an immediate need for IEEE P1900.5 policy languages. The paper shows the current lack of consensus on the scope and needs of policy languages, tracing roots to differences am...

This paper proposes an approach to integrate the self-organizing map (SOM) and kernel density estimation (KDE) techniques for the anomaly-based network intrusion detection (ABNID) system to monitor the network traffic and capture potential abnormal behaviors. With the continuous development of network technology, information security has become a m...