Sink-oriented tree based data dissemination protocol for mobile sinks wireless sensor networks

Wireless Networks - Data dissemination toward static sinks causes the nearby nodes to deplete their energy quicker than the other nodes in the field (i.e., this is referred to as the hotspot...     

Abnormal event detection and localization in crowded scenes based on PCANet

Detecting and localizing abnormal events in crowded scenes still remains a challenging task among computer vision community. An unsupervised framework is proposed in this paper to address the problem. Low-level features and optical flows (OF) of video sequences are extracted to represent motion information in the temporal domain. Moreover, abnormal events usually occur in local regions and are closely linked to their surrounding areas in the spatial domain. To extract high-level information from local regions and model the relationship in spatial domain, the first step is to calculate optical flow maps and divide them into a set of non-overlapping sub-maps. Next, corresponding PCANet models are trained using the sub-maps at same spatial location in the optical flow maps. Based on the block-wise histograms extracted by PCANet models, a set of one-class classifiers are trained to predict the anomaly scores of test frames. The framework is completely unsupervised because it utilizes only normal videos. Experiments were carried out on UCSD Ped2 and UMN datasets, and the results show competitive performance of this framework when compared with other state-of-the-art methods.     

A Novel Two-stage Learning Pipeline for Deep Neural Networks

In this work, a training method was proposed for Deep Neural Networks (DNNs) based on a two-stage structure. Local DNN models are trained in all local machines and uploaded to the center with partial training data. These local models are integrated as a new DNN model (combination DNN). With another DNN model (optimization DNN) connected, the combination DNN forms a global DNN model in the center. This results in greater accuracy than local DNN models with smaller amounts of data uploaded. In this case, the bandwidth of the uploaded data is saved, and the accuracy is maintained as well. Experiments are conducted on MNIST dataset, CIFAR-10 dataset and LFW dataset. The results show that with less training data uploaded, the global model produces greater accuracy than local models. Specifically, this method focuses on condition of big data.     

Single sample per person face recognition with KPCANet and a weighted voting scheme

Most current methods of facial recognition rely on the condition of having multiple samples per person available for feature extraction. In practical applications, however, only one sample may be available for each person to train a model with. As a result, many of the traditional methods fall short, leaving the challenge of facial recognition greater than ever. To deal with this challenge, this study addresses a face recognition algorithm based on a kernel principal component analysis network (KPCANet) and then proposes a weighted voting method. First, the aligned face image is partitioned into several non-overlapping patches to form the training set. Next, a KPCANet is used to obtain filters and feature banks. Finally, the identification of the unlabeled probe occurs through the application of the weighted voting method. Based on several widely used face datasets, the results of the experiments demonstrate the superiority of the proposed method.