• Journal of Electrical Engineering and Technology
• /
• v.11 no.6
• /
• pp.1857-1862
• /
• 2016

Human activity recognition using depth information is an emerging and challenging technology in computer vision due to its considerable attention by many practical applications such as smart home/office system, personal health care and 3D video games. This paper presents a novel framework of 3D human body detection, tracking and recognition from depth video sequences using spatiotemporal features and modified HMM. To detect human silhouette, raw depth data is examined to extract human silhouette by considering spatial continuity and constraints of human motion information. While, frame differentiation is used to track human movements. Features extraction mechanism consists of spatial depth shape features and temporal joints features are used to improve classification performance. Both of these features are fused together to recognize different activities using the modified hidden Markov model (M-HMM). The proposed approach is evaluated on two challenging depth video datasets. Moreover, our system has significant abilities to handle subject's body parts rotation and body parts missing which provide major contributions in human activity recognition.

• Smart Structures and Systems
• /
• v.20 no.1
• /
• pp.99-114
• /
• 2017

In cognitive science, it is illustrated how the collective opinions of a group of individuals answers to questions involving quantity estimation. One example of this approach is introduced in this article as Star Graph (SG) algorithm. This graph describes the details of communication among individuals to share their information and make a new decision. A new labyrinthine network of neighbors is defined in the decision-making process of the algorithm. In order to prevent getting trapped in local optima, the neighboring networks are regenerated in each iteration of the algorithm. In this algorithm, the normal distribution is utilized for a group of agents with the best results (guidance group) to replace the existing infeasible solutions. Here, some new functions are introduced to provide a high convergence for the method. These functions not only increase the local and global search capabilities but also require less computational effort. Various benchmark functions and engineering problems are examined and the results are compared with those of some other algorithms to show the capability and performance of the presented method.

• Structural Engineering and Mechanics
• /
• v.60 no.3
• /
• pp.471-488
• /
• 2016

This study predicts the strength of rotary brace damper by analyzing a new set of probabilistic models using the usual method of multiple linear regressions (MLR) and advanced machine-learning methods of multivariate adaptive regression splines (MARS), Rotary brace damper can be easily assembled with high energy-dissipation capability. To investigate the behavior of this damper in structures, a steel frame is modeled with this device subjected to monotonic and cyclic loading. Several response parameters are considered, and the performance of damper in reducing each response is evaluated. MLR and MARS methods were used to predict the strength of this damper. Displacement was determined to be the most effective parameter of damper strength, whereas the thickness did not exhibit any effect. Adding thickness parameter as inputs to MARS and MLR models did not increase the accuracies of the models in predicting the strength of this damper. The MARS model with a root mean square error (RMSE) of 0.127 and mean absolute error (MAE) of 0.090 performed better than the MLR model with an RMSE of 0.221 and MAE of 0.181.

• Smart Structures and Systems
• /
• v.14 no.6
• /
• pp.1055-1079
• /
• 2014

Substructure shake table testing is a class of real-time hybrid simulation (RTHS). It combines shake table tests of substructures with real-time computational simulation of the remaining part of the structure to assess dynamic response of the entire structure. Unlike in the conventional hybrid simulation, substructure shake table testing imposes acceleration compatibilities at substructure boundaries. However, acceleration tracking of shake tables is extremely challenging, and it is not possible to produce perfect acceleration tracking without time delay. If responses of the experimental substructure have high correlation with ground accelerations, response errors are inevitably induced by the erroneous input acceleration. Feeding the erroneous responses into the RTHS procedure will deteriorate the simulation results. This study presents a set of techniques to enable reliable substructure shake table testing. The developed techniques include compensation techniques for errors induced by imperfect input acceleration of shake tables, model-based actuator delay compensation with state observer, and force correction to eliminate process and measurement noises. These techniques are experimentally investigated through RTHS using a uni-axial shake table and three-story steel frame structure at the Johns Hopkins University. The simulation results showed that substructure shake table testing with the developed compensation techniques provides an accurate and reliable means to simulate the dynamic responses of the entire structure under earthquake excitations.

• Smart Structures and Systems
• /
• v.16 no.4
• /
• pp.725-742
• /
• 2015

Tuned mass dampers (TMD) are devices employed in vibration control since the beginning of the twentieth century. However, their implementation for controlling the seismic response in civil structures is more recent. While the efficiency of TMD on structures under far-field earthquakes has been demonstrated, the convenience of its employment against near-fault earthquakes is still under discussion. In this context, the study of this type of device is raised, not as an alternative to the seismic isolation, which is clearly a better choice for new buildings, but rather as an improvement in the structural safety of existing buildings. Seismic records with an impulsive character have been registered in the vicinity of faults that cause seismic events. In this paper, the ability of TMD to control the response of structures that experience inelastic deformations and eventually reach collapse subject to the action of such earthquakes is studied. The results of a series of nonlinear dynamic analyses are presented. These analyses are performed on a numerical model of a structure under the action of near-fault earthquakes. The structure analyzed in this study is a steel frame which behaves as a single degree of freedom (SDOF) system. TMD with different mass values are added on the numerical model of the structure, and the TMD performance is evaluated by comparing the response of the structure with and without the control device.

• Smart Structures and Systems
• /
• v.16 no.5
• /
• pp.807-833
• /
• 2015

While tuned mass dampers are found to be effective in suppressing vibration in a tall building, integrating it with a semi-active control system enables it to perform more efficiently. In this paper a forty-story tall steel-frame building designed according to the Canadian standard, has been studied with and without semi-active and passive tuned mass dampers. The building is assumed to be located in the Vancouver, Canada. A magneto-rheological fluid based semi-active tuned mass damper has been optimally designed to suppress the vibration of the structure against seismic excitation, and an appropriate control procedure has been implemented to optimize the building's semi-active tuned mass system to reduce the seismic response. Furthermore, the control system parameters have been adjusted to yield the maximum reduction in the structural displacements at different floor levels. The response of the structure has been studied with a variety of ground motions with low, medium and high frequency contents to investigate the performance of the semi-active tuned mass damper in comparison to that of a passive tuned mass damper. It has been shown that the semi-active control system modifies structural response more effectively than the classic passive tuned mass damper in both mitigation of maximum displacement and reduction of the settling time of the building.

• Smart Structures and Systems
• /
• v.16 no.6
• /
• pp.1049-1068
• /
• 2015

In the paper, the authors propose the application of operational deflection shapes (ODS) for the detection of structural changes in technical objects. The ODS matrix is used to formulate the spatial filter that is further used for damage detection as a classical modal filter (Meirovitch and Baruh 1982, Zhang et al. 1990). The advantage of the approach lies in the fact that no modal analysis is required, even on the reference spatial filter formulation and other components apart from structural ones can be filtered (e.g. harmonics of rotational velocity). The proposed methodology was tested experimentally on a laboratory stand, a frame-like structure, excited from two sources: an impact hammer, which provided a wide-band excitation of all modes, and an electro-dynamic shaker, which simulated a harmonic component in the output spectra. The damage detection capabilities of the proposed method were tested by changing the structural properties of the model and comparing the results with the original ones. The quantitative assessment of damage was performed by employing a damage index (DI) calculation. Comparison of the output of the ODS filter and the classical modal filter is also presented and analyzed in the paper. The closing section of the paper describes the verification of the method on a real structure - a road viaduct.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2767-2780
• /
• 2016

Video-based human-activity recognition has become increasingly popular due to the prominent corresponding applications in a variety of fields such as computer vision, image processing, smart-home healthcare, and human-computer interactions. The essential goals of a video-based activity-recognition system include the provision of behavior-based information to enable functionality that proactively assists a person with his/her tasks. The target of this work is the development of a novel approach for human-activity recognition, whereby human-body-joint features that are extracted from depth videos are used. From silhouette images taken at every depth, the direction and magnitude features are first obtained from each connected body-joint pair so that they can be augmented later with motion direction, as well as with the magnitude features of each joint in the next frame. A generalized discriminant analysis (GDA) is applied to make the spatiotemporal features more robust, followed by the feeding of the time-sequence features into a Hidden Markov Model (HMM) for the training of each activity. Lastly, all of the trained-activity HMMs are used for depth-video activity recognition.

• International Journal of Contents
• /
• v.10 no.1
• /
• pp.1-11
• /
• 2014

In the current generation of smart mobile devices, object tracking is one of the most important research topics for computer vision. Because human face tracking can be widely used for many applications, collecting a dataset of face videos is necessary for evaluating the performance of a tracker and for comparing different approaches. Unfortunately, the well-known benchmark datasets of face videos are not sufficiently diverse. As a result, it is difficult to compare the accuracy between different tracking algorithms in various conditions, namely illumination, background complexity, and subject movement. In this paper, we propose a new dataset that includes 91 face video clips that were recorded in different conditions. We also provide a semi-automatic ground-truth generation tool that can easily be used to evaluate the performance of face tracking systems. This tool helps to maintain the consistency of the definitions for the ground-truth in each frame. The resulting video data set is used to evaluate well-known approaches and test their efficiency.

• Smart Structures and Systems
• /
• v.14 no.5
• /
• pp.831-845
• /
• 2014

This paper proposed a discrete wavelet transform based method for time-varying physical parameter identification of shear type structures. The time-varying physical parameters are dispersed and expanded at multi-scale as profile and detail signal using discrete wavelet basis. To reduce the number of unknown quantity, the wavelet coefficients that reflect the detail signal are ignored by setting as zero value. Consequently, the time-varying parameter can be approximately estimated only using the scale coefficients that reflect the profile signal, and the identification task is transformed to an equivalent time-invariant scale coefficient estimation. The time-invariant scale coefficients can be simply estimated using regular least-squares methods, and then the original time-varying physical parameters can be reconstructed by using the identified time-invariant scale coefficients. To reduce the influence of the ill-posed problem of equation resolving caused by noise, the Tikhonov regularization method instead of regular least-squares method is used in the paper to estimate the scale coefficients. A two-story shear type frame structure with time-varying stiffness and damping are simulated to validate the effectiveness and accuracy of the proposed method. It is demonstrated that the identified time-varying stiffness is with a good accuracy, while the identified damping is sensitive to noise.