Xiaojuan Wu

An Adaptive Cross-Layer Scheduling Algorithm for Multimedia Networks

Nowadays wireless networks standards, e.g. IEEE 802.16, aim to accommodate diverse QoS requirements for multimedia users. The standard provided four kinds of multimedia data services, associated with QoS parameters and left the QoS scheduling algorithm undefined. In this paper, we propose an adaptive cross physical (PHY) layer and MAC layer scheduling algorithm for IEEE 802.16 BWA system. The proposed scheduler employs adaptive modulation and coding (AMC) scheme at the PHY layer according to the signal to noise ratio (SNR) on wireless fading channels. And we also define each kind of multimedia connections the cost function based on its service status, throughout or deadline in MAC layer. The simulation results show that our proposed scheduler can provide diverse QoS guarantees, maximum throughput and achieve an optimum tradeoff between throughput and fairness.

Automatic Detection and Tracking of Maneuverable Birds in Videos

In this paper, we try to detect and track maneuverable birds in captured videos for further automatic research. To avoid abrupt scene change, videos are captured by pointing the camera upwards and making the image against the sky. Two levels (pixel level and frame level) background update algorithm is used to get the foreground in real time. In frame level, three thresholds are used to update the background to achieve fast background update under abrupt scene changing. Targets are detected from the binary foreground after retrieving contours, size filter and margin measure. The tracking of maneuverable birds is achieved by using a Markov chain Monte Carlo (MCMC) filter with no move types. The experiment results show that multiple maneuverable birds are detected and tracked accurately in real time, and the size of the tracking box adjust fast to cover the true area of birds.

Wavelet-Based Far Infrared Medical Image Compression with Histogram Shifting

A near-lossless compression algorithm for far infrared medical image is proposed based on lifting wavelet transform and histogram shifting. The algorithm can losslessly compress the data in the human body area with integrating the source data and normalizing the useless background (lossy). Experiment results show that it is an effective algorithm compared with JPEG-LS and can compress the image more than 6 times without affecting clinical diagnostics.

Robust post-processing strategy for gait silhouette

An integrated silhouette with perfect appearance is helpful for gait recognition. However, the silhouettes often have holes or missing parts, because the commonly used motion detection and extraction methods are not always suitable for every case. A robust post-processing strategy is proposed here to refine the raw silhouettes. First, an individual silhouette model which represents the mutual characteristic of a certain sequence is trained to update the current sequence. Then a population silhouette model which represents the mutual characteristic of all sequences in the dataset is trained to update the sequences that need further refinement. The experiments on NLPR database show that the proposed algorithm is quite effective and helps the existing recognition method to achieve higher classification performance.

A New Gait Analyzing Method Under Monocular Camera

View point is a key factor in the field of gait analysis. This paper put forward a novel gait analyzing method which can avoid the affection of view points in some extent. It sponsors a new way of 3D point reconstruction from an image captured by a monocular camera with a coordinate conversion factor from the view of projective geometry through parallel restriction. In gait analysis, the human body is modeled by three connected sticks. Then we make use of the coordinate conversion factor to obtain the gait feature composed of five elements represented by the parameters of that model. In the experiments, our algorithm is tested by estimating the trajectory of walking people. The results reveal that the reconstruction is simple and effective. Moreover, the extracted features can partially avoid the affect of view points.

Gait Recognition Based on Fast Shape Matching

This paper proposes a novel gait recognition method based on fast shape matching algorithm. First, we propose a simple but effective gait cycle estimating method and analyze the periodic property of gait sequences. Each frame in the sequence is assigned a phase. Those frames with the same phase are averaged to clean noises. As a result, the original long sequence with dozens of frames is simply represented by a shorter representative sequence with half of cycle length. Shape context descriptor is introduced to depict the distribution of the boundary point of each frame. In order to decrease the computational cost of point matching, we propose a fast shape matching method. The computational efficiency is at least upgraded by two orders of magnitude. The experiments on NLPR database verify the effectiveness of the proposed method. Moreover, we also discuss the influence of the number of the most similar frame and the number of sample point.

New method of 3D point reconstruction from monocular camera

Put forward a method of 3D point reconstruction from an image taken by monocular camera. In algorithm, the coordinate conversion factor of the given point can be calculated through a pair of parallel lines, whose positions in the scene are known. Then reconstruction of any points in 3D space which contains parallel lines can be achieved using the coordinate conversion factor. In experiments, the algorithm is used to estimate the trajectory of walking people.

Birds’ Flap Frequency Measure Based on Automatic Detection and Tracking in Captured Videos

Flap frequency of birds, as an important feature in identification and behavior assessment to attractants, has not found a good way to measure in videos. In the paper, we try to measure flap frequency of birds automatically based on detection and tracking. The size of bird’s area in image fluctuates with wing flaps. In the 90° pitch angle case, when wings are flat, the area is the largest, and then the area decreases along with wings moving upwards or downwards. Based on the feature, we get flap frequency of birds by analyzing fluctuation periods of area size signals.

A dynamic bandwidth allocation scheme for EPON

This paper analyses current bandwidth schemes and proposes a novel dynamic bandwidth allocation scheme for EPON. According the scheme, we define four kinds of multimedia services such as Unsolicited Request Service (URS), Realtime Service (rt-S), Non-Real-time Service (nrt-S) and Best Effort (BE). Different kinds of services have different Quality of Service (QoS) requirements. Our scheme considers the diverse QoS request, e.g., delay for rt-S, throughput for nrt-S and fairness for BE. The simulation results show this novel scheme can ensure the quality of service (QoS) and improve bandwidth utilization.