Content-Based Image Retrieval (CBIR) is an important research topic of information retrieval,
involved in computer graphics, image processing, data mining and pattern recognizing. To make
content-based image retrieval suitable large-scale image database, we develop an effective dynamic
hierarchical clustering index scheme. Although this system uses a hierarchical clustering technology, with
the increasing in the number of cluster centers, it is slow to find the centers, and it becomes a system
performance bottleneck. In this paper, content features of image memory indexing is built. This method
effectively improves the retrieval speed without loss of the precision. Moreover, the clustering model was
improved, integrating the content features and textual features of image, which greatly improve the
accuracy of the clustering, thus significantly improves the system precision.
We propose a unified approach that incorporates the mean shift-based image segmentation algorithm and the SST (shortest spanning tree)-minmax-based graph grouping method to achieve effective IR object segmentation performance amenable for real-time application. It preprocesses an image by using the mean shift algorithm to form segmented regions that can not only remove the noise, but also preserve the desirable discontinuity characteristics of the ship object. The segmented regions can then effectively represent the original image by using the graph structures, and we apply the SST-minmax method to perform merging procedure to form the final segmented regions. Due to the good discontinuity-preserving filtering characteristic, we can effectively remove the clutter disturbance of the sea background without loss of the IR ship object information, and significantly reduce the number of basic image entities. Therefore, the region merging based on SST-minmax can produce excellent segmentation performance at low computational cost due to smaller clutter disturbance and less region nodes. The superiority of the proposed method is examined and demonstrated through a large number of experiments using a real IR ship image sequence.
Providing massive media streaming services is a very difficult problem for academic research and also a very constructive proposal for commercial investment. Traditional client/server model with very high cost/profit ratio and big bottlenecks, and novel P2P model with complex management, are all not good. In this paper we propose a semi-autonomous peer-to-peer-like network to provide massive media streaming services and avoid above problems. Our scheme is a tradeoff between server-systems and P2P systems. In this scheme, we make full use of P2P systems advantages, high aggregation network bandwidth and large storage spaces. But to avoid the impacts from peers unbending actions, we divide whole network serving area into many small zones and designate a seeding server for whole network area and a sub-seeding server for each network zone. Simulation results prove the advantages of the new system with good scalability and stable QoS.
An admission control algorithm is a very important component in a video server that can services large number clients simultaneously. Given the real-time requirements of each client and the fixed data transfer bandwidth, a VOD server must employ admission control algorithms to determine whether or not a new request can be admitted to the server without violating the requirements of the clients already being serviced. All prior admission control policies for VOD servers have focused on improving the storage system utilization such that a video server can service more users while satisfying the QoS requirements. For VOD service, the users can only view the video through transmission networks, so the transmission of video stream in the transmission networks is also very important. The admission control scheme presented in this paper is divided into two layers: disk subsystem admission control and network subsystem admission control. Through simulation, our admission control algorithm outperforms others significantly.
Detection rate is vital to intrusion detection. We propose a new search algorithm base on probability to speed up the process rate for a novel <i>compound intrusion detection system </i>(CIDS). We employ an improved Bayesian decision theorem to build this compound model. The improved Bayesian decision theorem brings four profits to this model. The first is to eliminate the flaws of a narrow definition for normal patterns and intrusion patterns. The second is to extend the known intrusions patterns to novel intrusions patterns. The third is to reduce risks that detecting intrusion brings to system. The last is to offer a method to build a compound intrusion detection model that integrates <i>misuse intrusion detection system </i>(MIDS) and <i>anomaly intrusion detection system </i>(AIDS). During the experiment of this model, we find that different system calls sequences have different probabilities. So the sequences with high probabilities are compared prior to an observed sequence, which is the foundation of our new search algorithm. We evaluate the performance of the new algorithm using numerical results, and the results show this new algorithm increases the detection rate.