Earticle

In this paper, all phase biorthogonal transform (APBT) based on parallel algorithm is proposed. It can solve two problems. First, block-based DCT transform coding has serious blocking artifacts when the image is highly compressed at low bit rate. Second, APBT can solve the problem about blocking artifacts, but it does not have a fast algorithm, it has a low efficiency when APBT applies to image processing. So APBT based on parallel algorithm can solve the above problems, and it provides more space for improving the processing speed of APBT. We use the CUDA toolkit based on GPU which is released by NVIDIA to design the parallel algorithm of APBT. Experimental results show that the maximum speedup ratio of parallel algorithm of APBT can reach more than 40 times with a very low version GPU, compared with conventional serial APBT. And the reconstructed image using the proposed algorithm has the same performance with the serial one in terms of objective quality and subjective effect. The proposed parallel algorithm based on GPU of APBT also can be used in image compression, video compression, the edge detection, and some other fields of image processing.

The purpose of this study is, in the situation that consumers use not only online news but social media as information delivery media, based on web data, to examine the effect of touch point activities in two media different according to product types on consumer search activities. To do this, we traced the patterns of information diffusion according to time, by extracting the number of the web postings of the information about the products from on-line news and blogs on a weekly or monthly basis, and identified consumer search activities about the goods by using their search traffics provided on a weekly basis by portal sites As a result of the study, it turned out that touch point activities and consumer search activities are different from each other in two media different according to product types. Specifically speaking, as for utilitarian goods, it appeared that there were correlations between the touch points and consumer search activities of the two different information media. In particular, more similarity in the patterns of information diffusion was found in the news sites than in the blogs. In the case of hedonic goods, the correlations between touch points and consumer search activities showed different results according to the time intervals. A monthly analysis showed that the consumer search activities had correlations not with on-line news, but with blogs. On the contrary, a weekly analysis showed that they had higher correlations with blogs than with on-line news sites though having correlations with both the two.

As a new learning framework, Multi-Instance learning is labeled recently and has successfully found application in vision classification. A novel Multi-instance bag generating method is presented in this paper on basis of Gaussian Mixed Model. The generated GMM model composes not only color but also the locally stable unchangeable components. It is frequently named as MI bag by researchers. Besides, another method called Agglomerative Information Bottleneck clustering is applied to replace the MIL problem with the help of single-instance learning ones. Meanwhile, single-instance classifiers are employed for classification. Finally, ensemble learning is adopted to strengthen classifiers’ generalization ability of RBM (Restricted Boltzmann Machine) as the base classifier. On the basis of large-scale datasets, this method is tested and the corresponding result shows that our method provides high accuracy and good performance for image annotation, feature matching and example-based object-classification.

Wireless sensor networks are being applied to a wider and wider range and better quality of network service is becoming increasingly important. Node deployment has been a key point of research on development of wireless sensor network. In essence, the problem node deployment of wireless sensor networks calculates the position where sensors are placed in regions to satisfy specific demands of network. This problem has been proved to be a NP-complete one. This thesis proposes node deployment algorithm about wireless sensor networks based on Steiner tree algorithm. On the basis of the triangle Steiner tree algorithm, it considers the two aspects algorithm complexity and practical topology to design improvement strategies. The algorithm can ensure high arithmetic speed and better results to solve node deployment based on network connectivity.

In this paper, we compare well known interpolation methods such as nearest neighbor, bilinear, bicubic, triangle kernel, and Lagrangian interpolation method. Reconstruction errors from above interpolation methods are compared using test image. From the simulation results, it can be found that Lagrangian method outperforms all other upsampling methods. Visual performance comparison is provided by using two LC images.

Disaster comes up always in form of chain of events. These occurrence and evolution are represented dynamically in our daily social media. Retrieving or tracing these user-oriented relevant events around disasters in an aggregated results list through meta-search engine is a useful application. The definition and detection of events has been a hot topic in the study of natural language processing. We give our opinion in this paper, and we gathering the disaster event information via different search interface for an event chain releasing is important. This paper designed an event-triggered model for events detection. As the emergency requirement of government as well as individual activities, disaster events detection is discussed as an example in this paper. A meta-search engine interface model is constructed based on the open source project Carrot2. And a H-T-E (Hazard, Trigger and Event) query expansion approach is applied in our event triggered model. The experimental results show that the proposed method can bring about increased accuracy and extra clues not supplied by commercial search engines. The solution we applied can be useful for other information supervision tasks and some issues like crisis ontology construction etc. also triggered by the project for further study.

Congestion in Wireless Sensor Networks (WSN) get worse when there are multiple and random flows of data in which some have superior significance over the others requiring fidelity in terms of packet delivery, QoS, energy efficiency and throughput. In node-level, congestion leads to impairment of packets that obviously reduces the QoS. In this paper, we present a Cluster based congestion control with Rate Adjustment based on Priority (CRAP) protocol, which self-organizes the sensor nodes into clusters. These clusters monitor congestion in a proactive manner within its confined range which exchanges information among them and adjusts the traffic rate when one cluster has high priority flow over the other. This rate adjustment is based on the exchange of traffic rate estimate among the clusters that reduces packet re-transmissions and energy loss. Our simulation expedites system wide rate control resulting in good throughput, very low packet loss probability and delay that deals with multiple, random flows of data.

Wireless Communication network is an inescapable reality of our days, being part of curricular proposed by almost all institute. Effective learning and teaching of wireless communication however, require student to visualize and illustrate in understanding the technical and how the communication network is operates. This paper recommends a conceptual framework to overcome the inability student in learning wireless communication network course which is complex and difficult especially for novice student. The aim of the course is to teach student not only programming and hardware design but also basic team-work and project management skills in the field of wireless networks through the active learning using the tools that we proposed. In this paper we proposed the architecture named Wireless Kit (Wkits) as one of the support tools that help novice student in enhancing their understanding of learning wireless communication. This kits focus in wireless communication such as Bluetooth, Wi-Fi and FRID technology.

Aimed to the difficult problem of detection information of video advertising from the massive video data, the novelty detection method of the video advertising based on the least squares support vector machine (LS-SVM) algorithm is introduced after analyzing and studying the feature of video advertising. The detection of the abrupt change shot and the gradual change shot in video data is realized by using color histogram algorithm, and the classification of the shot is achieved by using the particle swarm optimization and least squares support vector machine algorithm. The experiments show that the method based on the color histogram algorithm and LS-SVM algorithm has better Precision and recall, and can realize the accuracy detection of the video advertising.