Engineering Science, School of

Receive updates for this collection

Hand Motion and Posture Recognition in a Network of Calibrated Cameras

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2017-10-31
Abstract: 

This paper presents a vision-based approach for hand gesture recognition which combines both trajectory and hand posture recognition. The hand area is segmented by fixed-range CbCr from cluttered and moving backgrounds and tracked by Kalman Filter. With the tracking results of two calibrated cameras, the 3D hand motion trajectory can be reconstructed. It is then modeled by dynamic movement primitives and a support vector machine is trained for trajectory recognition. Scale-invariant feature transform is employed to extract features on segmented hand postures, and a novel strategy for hand posture recognition is proposed. A gesture vector is introduced to recognize hand gesture as an entirety which combines the recognition results of motion trajectory and hand postures where a support vector machine is trained for gesture recognition based on gesture vectors.

Document type: 
Article
File(s): 

Multimodal Sensing Interface for Haptic Interaction

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2017-04-04
Abstract: 

This paper investigates the integration of a multimodal sensing system for exploring limits of vibrato tactile haptic feedback when interacting with 3D representation of real objects. In this study, the spatial locations of the objects are mapped to the work volume of the user using a Kinect sensor. The position of the user’s hand is obtained using the marker-based visual processing. The depth information is used to build a vibrotactile map on a haptic glove enhanced with vibration motors. The users can perceive the location and dimension of remote objects by moving their hand inside a scanning region. A marker detection camera provides the location and orientation of the user’s hand (glove) to map the corresponding tactile message. A preliminary study was conducted to explore how different users can perceive such haptic experiences. Factors such as total number of objects detected, object separation resolution, and dimension-based and shape-based discrimination were evaluated. The preliminary results showed that the localization and counting of objects can be attained with a high degree of success. The users were able to classify groups of objects of different dimensions based on the perceived haptic feedback.

Document type: 
Article
File(s): 

Vertical Ground Reaction Force Marker for Parkinson’s Disease

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2017-05-11
Abstract: 

Parkinson’s disease (PD) patients regularly exhibit abnormal gait patterns. Automated differentiation of abnormal gait from normal gait can serve as a potential tool for early diagnosis as well as monitoring the effect of PD treatment. The aim of current study is to differentiate PD patients from healthy controls, on the basis of features derived from plantar vertical ground reaction force (VGRF) data during walking at normal pace. The current work presents a comprehensive study highlighting the efficacy of different machine learning classifiers towards devising an accurate prediction system. Selection of meaningful feature based on sequential forward feature selection, the swing time, stride time variability, and center of pressure features facilitated successful classification of control and PD gaits. Support Vector Machine (SVM), K-nearest neighbor (KNN), random forest, and decision trees classifiers were used to build the prediction model. We found that SVM with cubic kernel outperformed other classifiers with an accuracy of 93.6%, the sensitivity of 93.1%, and specificity of 94.1%. In comparison to other studies, utilizing same dataset, our designed prediction system improved the classification performance by approximately 10%. The results of the current study underscore the ability of the VGRF data obtained non-invasively from wearable devices, in combination with a SVM classifier trained on meticulously selected features, as a tool for diagnosis of PD and monitoring effectiveness of therapy post pathology.

Document type: 
Article
File(s): 

Classifying Three Imaginary States of the Same Upper Extremity Using Time-Domain Features

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2017-03-30
Abstract: 

Brain-computer interface (BCI) allows collaboration between humans and machines. It translates the electrical activity of the brain to understandable commands to operate a machine or a device. In this study, we propose a method to improve the accuracy of a 3-class BCI using electroencephalographic (EEG) signals. This BCI discriminates rest against imaginary grasps and elbow movements of the same limb. This classification task is challenging because imaginary movements within the same limb have close spatial representations on the motor cortex area. The proposed method extracts time-domain features and classifies them using a support vector machine (SVM) with a radial basis kernel function (RBF). An average accuracy of 74.2% was obtained when using the proposed method on a dataset collected, prior to this study, from 12 healthy individuals. This accuracy was higher than that obtained when other widely used methods, such as common spatial patterns (CSP), filter bank CSP (FBCSP), and band power methods, were used on the same dataset. These results are encouraging and the proposed method could potentially be used in future applications including BCI-driven robotic devices, such as a portable exoskeleton for the arm, to assist individuals with impaired upper extremity functions in performing daily tasks.

Document type: 
Article
File(s): 

White Matter fMRI Activation Cannot Be Treated as a Nuisance Regressor: Overcoming a Historical Blind Spot

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2019-10-04
Abstract: 

Despite past controversies, increasing evidence has led to acceptance that white matter activity is detectable using functional magnetic resonance imaging (fMRI). In spite of this, advanced analytic methods continue to be published that reinforce a historic bias against white matter activation by using it as a nuisance regressor. It is important that contemporary analyses overcome this blind spot in whole brain functional imaging, both to ensure that newly developed noise regression techniques are accurate, and to ensure that white matter, a vital and understudied part of the brain, is not ignored in functional neuroimaging studies.

Document type: 
Article
File(s): 

Application of Modified PageRank Algorithm for Anomaly Detection in Movements of Older Adults

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2019-03-11
Abstract: 

It is a well-known statistic that the percentage of our older adult population will globally surpass the other age groups. A majority of the elderly would still prefer to keep an active life style. In support of this life style, various monitoring systems are being designed and deployed to have a seamless integration with the daily living activities of the older adults while preserving various levels of their privacy. Motion tracking is one of these health monitoring systems. When properly designed, deployed, integrated, and analyzed, they can be used to assist in determining some onsets of anomalies in the health of elderly at various levels of their Movements and Activities of Daily Living (MADL). This paper explores how the framework of the PageRank algorithm can be extended for monitoring the global movement patterns of older adults at their place of residence. Through utilization of an existing dataset, the paper shows how the movement patterns between various rooms can be represented as a directed graph with weighted edges. To demonstrate how PageRank can be utilized, a base graph representing a normal pattern can be defined as what can be used for further anomaly detection (e.g., at some instances of observation the measured movement pattern deviates from what is previously defined as a normal pattern). It is shown how the PageRank algorithm can detect simulated change in the pattern of motion when compared with the base-line normal pattern. This feature can offer a practical approach for detecting anomalies in movement patterns associated with older adults in their own place of residence and in support of aging in place paradigm.

Document type: 
Article
File(s): 

Cell-Net: Embryonic Cell Counting and Centroid Localization via Residual Incremental Atrous Pyramid and Progressive Upsampling Convolution

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2019-06-05
Abstract: 

In-vitro fertilization (IVF), as the most common fertility treatment, has never reached its maximum potentials. Systematic selection of embryos with the highest implementation potentials is a necessary step towards enhancing the effectiveness of IVF. Embryonic cell numbers and their developmental rate are believed to correlate with the embryo's implantation potentials. In this paper, we propose an automatic framework based on a deep convolutional neural network to take on the challenging task of automatic counting and centroid localization of embryonic cells (blastomeres) in microscopic human embryo images. In particular, the cell counting task is reformulated as an end-to-end regression problem that is based on a shape-aware Gaussian dot annotation to map the input image into an output density map. The proposed Cell-Net system incorporates two novel components, residual incremental Atrous pyramid and progressive up-sampling convolution. Residual incremental Atrous pyramid enables the network to extract rich global contextual information without raising the ‘grinding’ issue. Progressive up-sampling convolution gradually reconstructs a high-resolution feature map by taking into account short- and longrange dependencies. Experimental results confirm that the proposed framework is capable of predicting the cell-stage and detecting blastomeres in embryo images of 1

Document type: 
Article
File(s): 

Performance Evaluation of Techniques for Identifying Abnormal Energy Consumption in Buildings

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2019-05-08
Abstract: 

Energy consumption in buildings has steadily increased. Buildings consume more energy than necessary due to suboptimal design and operation. Apart from retro-fitting, not much can be done with the design of the existing building, but the operation of the building can be improved. Ignoring or failing to fix the faults can lead to problems like the higher cost in excess energy usage or premature component failure. At the same time understanding, identifying, and addressing abnormal energy consumption in buildings can lead to energy savings and detection of faulty appliances. This paper investigates two key challenges found in energy anomaly detection research: 1) the lack of labeled ground truth and 2) the lack of consistent performance accuracy metrics. In the first challenge, labeled ground truth is imperative for training and benchmarking algorithms to detect anomalies. In the second challenge, consistent performance accuracy metrics are crucial to quantifying how well algorithms perform against each other. There exists no publicly available energy consumption dataset with labeled anomaly events. Therefore, we propose two approaches that help in the automatic annotation of the ground truth data from publicly available datasets: a statistical approach for short-term data and a piecewise linear regression method for long-term data. We demonstrate these approaches using two publicly available datasets called Dataport (Pecan Street) and HUE. Using different existing accuracy metrics, we run a series of experiments on anomaly detection algorithms and discuss what metrics can be best used for consistent accuracy testing amongst researchers. In addition, while providing the source code, we also release an anomaly annotated dataset produced by this source code.

Document type: 
Article
File(s): 

Deep Attention Models for Human Tracking Using RGBD

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2019-02-13
Abstract: 

Visual tracking performance has long been limited by the lack of better appearance models. These models fail either where they tend to change rapidly, like in motion-based tracking, or where accurate information of the object may not be available, like in color camouflage (where background and foreground colors are similar). This paper proposes a robust, adaptive appearance model which works accurately in situations of color camouflage, even in the presence of complex natural objects. The proposed model includes depth as an additional feature in a hierarchical modular neural framework for online object tracking. The model adapts to the confusing appearance by identifying the stable property of depth between the target and the surrounding object(s). The depth complements the existing RGB features in scenarios when RGB features fail to adapt, hence becoming unstable over a long duration of time. The parameters of the model are learned efficiently in the Deep network, which consists of three modules: (1) The spatial attention layer, which discards the majority of the background by selecting a region containing the object of interest; (2) the appearance attention layer, which extracts appearance and spatial information about the tracked object; and (3) the state estimation layer, which enables the framework to predict future object appearance and location. Three different models were trained and tested to analyze the effect of depth along with RGB information. Also, a model is proposed to utilize only depth as a standalone input for tracking purposes. The proposed models were also evaluated in real-time using KinectV2 and showed very promising results. The results of our proposed network structures and their comparison with the state-of-the-art RGB tracking model demonstrate that adding depth significantly improves the accuracy of tracking in a more challenging environment (i.e., cluttered and camouflaged environments). Furthermore, the results of depth-based models showed that depth data can provide enough information for accurate tracking, even without RGB information.

Document type: 
Article
File(s):