Many-core SoC (MCSoC) design is one of chip design trends and challenges. Modeling and simulation for large scale MCSoC design is the key to get it right the first time. NoC is the most important component of future M...
详细信息
Many-core SoC (MCSoC) design is one of chip design trends and challenges. Modeling and simulation for large scale MCSoC design is the key to get it right the first time. NoC is the most important component of future MCSoC. A novel modeling tool, MCVP-NoC (Many-Core Virtual Platform with Networks-on-Chip support) was designed for large scale MCSoC modeling, and can be used in design space exploration, early software development and system verification. MCVP-NoC was built on systemC, OVP(Open Virtual Platform) and TLM2.0, integrated with Orion2.0 for power and area estimation. MCVP-NoC can run real software code, and can be used to evaluate the performance of the MCSoC design under real application load.
A method of spatial super-resolution was proposed based on the fuzzy registration. Firstly, the compensated frame that is blurred was obtained using optical flow and motion compensation methods. Then, pixels were divi...
详细信息
A method of spatial super-resolution was proposed based on the fuzzy registration. Firstly, the compensated frame that is blurred was obtained using optical flow and motion compensation methods. Then, pixels were divided into two classes, related pixels and non-related pixels, through correlation judgment strategy. Moreover, the local patch motion was analyzed. For simple motion patch, gray vector was used to represent the local features. For complex motion patch, in order to improve the registration accuracy, the Zernike-moment was calculated to characterize the patch. Experimental results show that the proposed method can not only compensate the fuzzy frame, but also improve the performance of reconstruction.
An image fusion algorithm based on SCDPT was proposed, which is under the background of the space docking. The image fusion rule was proposed based on matching degrees-integrated energy region, and applied to the meth...
详细信息
An image fusion algorithm based on SCDPT was proposed, which is under the background of the space docking. The image fusion rule was proposed based on matching degrees-integrated energy region, and applied to the method to the high-frequency coefficients created by SCDPT decomposition. The experimental results of subjective evaluation criterion as well as objective evaluation criterion show that the improved fusion rule for SCDPT algorithm can achieve satisfactory fusion results both with standard pictures and space pictures compared to the traditional wavelet method and UDCT.
Data availab.lity is a desirable feature for all cloud providers. However, data reliability and disaster recovery cost are always two incompatible goals because adoption of data redundancy will bring incremental cost....
详细信息
Frequent itemsets, also called frequent patterns, are important information about databases, and mining efficiently frequent itemsets is a core problem in data mining area. Pattern growth approaches, such as the class...
详细信息
In the Web Service world, it is an important technique that taking the usable behaviors in the existed web services provides some value-added functions by the service composition. Therefore, the usability analysis of ...
详细信息
LLE is a nonlinear dimensionality reduction method, which has been successfully applied to data visualization. Based on the assumption of local linearity, LLE can compute the weights between the KNN nodes using the lo...
详细信息
In this paper, we propose a deep packet inspection system based on the MapReduce. The MapReduce which is a parallel distributed programming model developed by Google applies the technology of deep packet inspection. N...
详细信息
A novel alive entropy-based detection approach was proposed, which detects the abnormal network traffic based on the values of alive entropies. The alive entropies calculated based on the NetFlow data coming from the ...
详细信息
A novel alive entropy-based detection approach was proposed, which detects the abnormal network traffic based on the values of alive entropies. The alive entropies calculated based on the NetFlow data coming from the network traffic of input and output of a whole system, which is essentially a monitored network. In order to decrease false positive rate of abnormal network traffic, different scales are selected to compute the values of alive entropies in different sizes of network traffic. With the low false positive rate of abnormal network traffic, the abnormal network traffic can be effectively detected. Experiments carried out on a real campus network were used to evaluate the effectiveness of the proposed approach. A comparative study illustrates that the proposed approach may easily detect the abnormal network traffic with random characteristics in comparison with some "conventional" approaches reported in the literatures.
Gesture inputs on multi-touch tabletops usually involve multiple fingers (more than two) and casual touchdowns or liftoffs of fingers. This flexibility of touch gestures allows more natural user interaction, but also ...
详细信息
暂无评论