Pengfei Duan

Multitarget tracking of pedestrians in video sequences based on particle filters

Video target tracking is a critical problem in the field of computer vision. Particle filters have been proven to be very useful in target tracking for nonlinear and non-Gaussian estimation problems. Although most existing algorithms are able to track targets well in controlled environments, it is often difficult to achieve automated and robust tracking of pedestrians in video sequences if there are various changes in target appearance or surrounding illumination. To surmount these difficulties, this paper presents multitarget tracking of pedestrians in video sequences based on particle filters. In order to improve the efficiency and accuracy of the detection, the algorithm firstly obtains target regions in training frames by combining the methods of background subtraction and Histogram of Oriented Gradient (HOG) and then establishes discriminative appearance model by generating patches and constructing codebooks using superpixel and Local Binary Pattern (LBP) features in those target regions. During the process of tracking, the algorithm uses the similarity between candidates and codebooks as observation likelihood function and processes severe occlusion condition to prevent drift and loss phenomenon caused by target occlusion. Experimental results demonstrate that our algorithm improves the tracking performance in complicated real scenarios.

A localization algorithm in wireless sensor networks based on PSO

Node localization is a fundamental and important technology in wireless sensor networks. In this paper, a localization algorithm in wireless sensor networks based on PSO is proposed. Unlike most of the existing location algorithm, the proposed algorithm figures out the rectangular estimation range of unknown node by bounding box algorithm and takes one value as the estimated coordinates of this node, then it has been optimized by PSO, so got the more precise location of unknown nodes. Simulation results show that this optimized algorithm outperforms traditional bounding box on the positioning accuracy and localization error.

Space-time simulation model based on particle swarm optimization algorithm for stadium evacuation

In this paper, a space-time simulation model based on particle swarm optimization algorithm for stadium evacuation is presented. In this new model, the fast evacuation, going with the crowd and the panic behaviors are considered and the corresponding moving rules are defined. The model is applied to a stadium and simulations are carried out to analyze the spacetime evacuation efficiency by different behaviors. The simulation results show that the behaviors of going with the crowd and panic will slow down the evacuation process while quickest evacuation psychology can accelerate the process, and panic is helpful to some extent. The setting of parameters is discussed to obtain best performance. The simulation results can offer effective suggestions for evacuees under emergency situation.

Very Deep Neural Networks for Hindi/Arabic Offline Handwritten Digit Recognition

Handwritten Digit Recognition (HDR) has become one of the challenging areas of research in the field of document image processing during the last few decades. In this paper, inspired by the success of the very deep state-of-the-art VGGNet, we proposed VGG_No for HDR. VGG_No is fast and reliable, which improved the classification performance effectively. Besides, this model has also reduced the overall complexity of VGGNet. VGG_No constructed by thirteen convolutional layers, two max-pooling layers, and three fully connected layers. A Cross-Validation analysis has been performed using the 10-Fold Cross-Validation strategy, and 10-Fold classification accuracies of 99.57% and 99.69% have been obtained for ADBase database and MNIST database, respectively. The classification performance of VGG_No is superior to existing techniques using multi-classifiers since it has achieved better results using very simple and homogeneous architecture.

Multi-column Deep Neural Network for Offline Arabic Handwriting Recognition

In recent years Deep Neural Networks (DNNs) have been successfully applied to several pattern recognition filed. For example, Multi-Column Deep Neural Networks (MCDNN) achieve state of the art recognition rates on Chinese characters database. In this paper, we utilized MCDNN for Offline Arabic Handwriting Recognition (OAHR). Through several settings of experiments using the benchmarking IFN/ENIT Database, we show incremental improvements of the words recognition comparable to approaches used Deep Belief Network (DBN) or Recurrent Neural Network (RNN.) Lastly, we compare our best result to those of previous state-of-the-arts.

Multi-objective optimization model based on steady degree for teaching building evacuation

In this paper, the process of evacuation in teaching building is considered. The concept of steady degree based on cellular automata and potential field is introduced and it can describe the behavior tendency of evacuees during the evacuation process. With the help of steady degree, the model simulates the indoor evacuation behavior. To reduce the congestion and evacuation time, a multi-objective optimization model considering steady degree and evacuation clearance time is proposed. Finally, an experiment in the Teaching Building No.1 of Wuhan University of Technology is carried out. The results show that this model can reduce the clearance time of emergency evacuation in teaching building compared to other models.

Hybrid Particle Filter Algorithm Using Multi Features for Video Target Tracking

Video target tracking is an essential problem in the field of computer vision. Particle filters have been proved to be very useful in target tracking for non-linear and non-Gaussian estimation problems. However, for the target tracking in complex background, it is often difficult to achieve robust tracking by using single target feature information. To solve this problem, this paper presents a hybrid particle filter algorithm using multi features for video target tracking. The algorithm integrates multiple features into particle filter to get better observation results, and then automatically adjusts the weight value of each feature according to the current tracking environment. In order to describe the target movement well, the method automatically adjusts the transfer range of particles according to the target speed changes, thus the particles can reach high likelihood region, which reduces the target lost phenomenon caused by speed changes. Experimental results demonstrate that the proposed algorithm improves the tracking performance in complicated real scenarios.

PSO-based emergency evacuation simulation

The Emergency Evacuation Simulation (EES) has been increasingly becoming a hotspot in the field of transportation. PSO-based EES is a good choice as its low computation complexity compared with some other algorithms, especially in an emergency. The selection of fitness function of each particle in PSO is a key problem for EES. This paper will introduce some fitness functions for EES and present a new fitness function called Triple-Distance Safe Degree (TDSD). Through theoretical analysis and experimental validation, the TDSD is proved to be much better than other fitness functions introduced in this paper.