Jiang Lu

Permanent URI for this collectionhttps://hdl.handle.net/10657.1/2411

Image

Dr. Jiang Lu is an Assistant Professor of Computer Engineering at University of Houston-Clear Lake. Dr. Lu's research interest draws in all aspects of intelligent sensor systems, including wireless communication, intelligent algorithms and the establishment of biomedical applications with intelligent sensor systems.


Browse

Recent Submissions

Now showing 1 - 10 of 10
  • Item
    Distributed Multiple Human Tracking with Wireless Binary Pyroelectric Infrared (PIR) Sensor Networks
    (IEEE, 2010) Lu, Jiang
    This paper presents a distributed multiple human tracking system based on binary pyroelectric infrared (PIR) sensors. The goal of our research is to make wireless distributed pyroelectric sensors a low-cost, low-data throughput alternative to the expensive infrared video sensors in surveillance applications. With the help of coded Fresnel lens arrays, a binary pyroelectric sensor array can easure the angular displacements of up to two thermal targets. The distributed multiple target tracking scheme is achieved by using (1) joint probabilistic association and (2) consensus filtering. The former can facilitate each sensor node to fuse the measurements and states of nodes within its neighborhood. The latter can guarantee that a consensus will be achieved among those distributed sensor nodes. A prototype wireless pyroelectric sensor network system has been developed to demonstrate the scalability and performance of the proposed distributed multiple human tracking system.
  • Item
    Space Encoding Based Compressive Multiple Human Tracking with Distributed Binary Pyroelectric Infrared Sensor Networks
    (IEEE, 2012) Lu, Jiang
    This paper presents a distributed, compressive multiple human tracking system based on binary pyroelectric infrared (PIR) sensor networks. The goal of our research is to develop an energy-efficient, low-data-throughput infrared surveillance system for various indoor applications. The compressive measurements are achieved by using techniques of (1) multiplex binary sensing and (2) space encoding. The target positions are reconstructed from the binary compressive measurements through (1) an expectation-maximization (EM) framework for space decoding, (2) representing the prior knowledge of target/sampling geometries with statistical parameters, and (3) hierarchical space encoding/decoding for multiple targets tracking. A wireless networked PIR sensor is designed to demonstrate the improved sensing efficiency and system scalability of the proposed distributed multiple human tracking system. The proposed compressive tracking framework can be extended to various binary sensing modalities.
  • Item
    Multi-agent Based Wireless Pyroelectric Infrared Sensor Network for Multi-human Tracking and Self-calibration
    (IEEE, 2013) Lu, Jiang
    This paper presents a multi-agent-based wireless pyroelectric infrared sensor network for human tracking and self-calibration. The goal of this research is to achieve a scalable, reconfigurable multi-agent system (MAS) which consists of sensing, action, decision agents can be developed from one uniform reconfigurable node, Sensing agents contain PIR sensors, a signal conditioning circuit, and programmable system on a chip (PSoC). Action agents contain a servo motor, a servo control circuit, and PSoC. Decision agents contain a field-programmable gate array (FPGA) board, which can implement self-calibration algorithm. Database agents are developed using SUN MySQL platform, which contains situation and group information (e.g., number of targets, system geometric parameters, nodes position, and orientations). Initial experimental results illustrate the advantage of the proposed MAS based PIR sensor networks.
  • Item
    Measuring Activities and Counting Steps with Smartsocks - an Unobtrusive and Accurate Method
    (IEEE, 2014) Lu, Jiang
    Physical inactivity is an important contributor to non-communicable diseases in countries of high income, and increasingly so in those of low and middle income. Physical inactivity is the leading cause of many diseases. It has been estimated that as many as 250,000 deaths per year in the United States, approximately 12% of the total, are attributable to a lack of regular physical activity. Measuring physical activities and counting steps is an effective method to diagnose some diseases. It can also serve as an effective method to encourage people to increase their physical activity. Pedometers have been invented as a convenient way of counting steps. However, most of them lack the functionality of differentiating activities. Pressure sensor pads can measure steps and gait, but as the pad has a limited size, it can not meet the need of anytime, anywhere usage. In this study, we made the Sensor Socks for measuring physical activities and counting steps. It is unobtrusive and convenient for everyday usage. Our experimental results show that the system has a high accuracy of the classification of physical activities and counting steps in a home or community environment.
  • Item
    Non-informative Hierarchical Bayesian Inference for Non-negative Matrix Factorization
    (Journal of Signal Processing, 2015) Lu, Jiang
    Non-negative matrix factorization (NMF) is an intuitive, non-negative, and interpretable approximation method. Canonical NMF approach could derive some basic components to represent original data, while probabilistic NMF approaches try to introduce some reasonable constraints to optimize the canonical NMF model. However, both of them cannot handle ground-truth bases discovering and model order determination problems. In general, the model order of basis matrix needs to be pre-defined. The model order determines the capability and accuracy of data structure discovering. However, how to accurately infer the model order of basis matrix has not been well investigated. In this paper, we propose a method called non-informative hierarchical Bayesian non-negative matrix factorization (NHBNMF) to automatically determine the model order and discover the data structure. They are achieved through hierarchical Bayesian inference model, maximum a posteriori (MAP) criterion, and non-informative parameters. In NHBNMF method, we first introduce a structure with two-level parameters to enable the entire model to approach the distribution of ground-truth bases. Then we use non-informative parameter scheme to eliminate the hyper-parameter to enable automatic searching. Finally, the model order and ground-truth bases are discovered by using MAP criterion and L2-norm selection. The experiments are conducted based on both synthetic and real-world datasets to show the effectiveness of our algorithm. The results demonstrate that our algorithm can accurately estimate the model order and discover the ground-truth bases. Even for the complicated FERET facial dataset, our algorithm still obtained interpretable bases an achieved satisfactory accuracy of the model order estimation.
  • Item
    Monitoring of Paces and Gaits Using Binary PIR Sensors with Rehabilitation Treadmill
    (IEEE, 2016) Lu, Jiang
    Recently, rehabilitation treadmills are designed for helping injured persons such as stroke patients and injured athletes in the process of physical therapy. By monitoring the changes of paces and gaits are wearable and/or expensive. This paper presents an inexpensive, non-intrusive wireless binary sensor system for pace estimation and lower-extremity gait recognition with low data throughput and high energy efficiency. The asymmetric but periodic movement of the injured person allows the study of pace and gait. The pace estimation is achieved by using the autocorrelation function. The gait information is represented by three features (1) temporal correlation, (2) marginal density (intersection probability), and (3) spatial correlation from binary data steam Experimental results show that our system can estimate the pace of walking or running with the accuracy of 97.7%. By using only three features, abnormal gaits can also be recognized.
  • Item
    Robot-assisted Intelligent Emergency System for Individual Elderly Independent Living
    (IEEE, 2016) Lu, Jiang
    With the ever-increasing aging population and others in need of close monitoring, there is a profound demand for an efficient and economical counter-emergency system. This paper proposes a smart connected real-time voice and video communication and tele-controlled co-robot system to provide a highly secured counter-emergency solution for people living alone, especially for elder adults. A comprehensive personal emergency assistant tele-robotics guard system is provided pervasive emergency intelligence for independent living. It contains three modules: robot awareness module, wearable emergency analysis module, and remote-control module. The robot in the robot awareness module is able to find out the yell of "help". The wearable emergency analysis module can recognize a person's activity and predict health condition. The remote-control module can control the robot remotely by using mobile devices. This smart connected tele-robotics guard system will dynamically detect, alarm, tele-operate for emergency situations. The prototypes have designed and developed with vocal help requests and wearable sensors to actively protect a person in emergency situations even when he/she loses consciousness, mobility, or clear conversation capability. The latest prototype was deployed in a senior home facility for extensive testing, eliciting real-life data to conduct further improvements.
  • Item
    A Smart System for Face Detection with Spatial Correlation Improvememt in IoT Environment
    (IEEE Smartworld, 2017) Lu, Jiang
    This paper presents an Internet of Things (IoT) based face detection system. The main objective is to build a fully automated human face detection system for images with complex backgrounds. Our system can capture images and run face detection program code. The Raspberry Pi 3 board is used as the gateway of IoT smart camera device with a camera on it. The final detected faces will be sent to remote devices, e.g. cell phones and laptops. Two main steps are used to detect faces: 1) cascade classifiers for face detection first and then 2) spatial correlation for detection results improvement. The experimental results showed that our smart camera system gives comparable face detection performance. The average precision of 85.7%. Several promising directions for future research based on IoT are also concluded.
  • Item
    Preprocessing Design in Pyroelectric Infrared Sensor Based Human Tracking System: On Sensor Selection and Calibration
    (IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2016) Lu, Jiang
    This paper presents an information-gain-based sensor selection approach as well as a sensor sensing probability model-based calibration process for multi human tracking in distributed binary pyroelectric infrared sensor networks. This research includes three contributions: 1) choose the subset of sensors that can maximize the mutual information between sensors and targets; 2) find the sensor sensing probability model to represent the sensing space for sensor calibration; 3) provide a factor graph-based message passing scheme for distributed tracking. Our approach can find the solution for sensor selection to optimize the performance of tracking. The sensing probability model is efficiently optimized through the calibration process in order to update the parameters of sensor positions and rotations An application for mobile calibration and tracking is developed. Simulation and experimental results are provided to validate the proposed framework.
  • Item
    Binary Compressive Tracking
    (IEEE Transactions on Aerospace and Electronic Systems, 2017) Lu, Jiang
    This paper presents a compressive tracking framework using distributed binary sensors. The goal of this research is to achieve the minimum data throughput for an accurate multitarget tracking system through novel spatial sampling schemes. The framework consists of two main components: space encoding and measurement decoding. The space encoding scheme is based on the low-density parity-check matrix, which converts k-sparse target position vectors into different codewords. The measurement decoding scheme contains linear-programming-based localization and graphical-model-based tracking algorithms, which converts codewords into the states of multiple targets A posterior Cramer-Rao bound analysis is utilized to achieve the tradeoff between the compression ratio of measurements and the accuracy of the tracking system. Simulation and experimental results are provided to validate the proposed framework.