Traditional feature selection algorithm cannot process streaming feature data, the redundancy calculation is complicated and the description of the instance is not accurate enough. A multi-label Distribution learning Feature Selection with Streaming Data Using Rough Set (FSSRS) was proposed to solve the above problem. Firstly, the online streaming feature selection framework was introduced into multi-label learning. Secondly, the original conditional probability was replaced by the dependency in rough set theory, which made the streaming data feature selection algorithm more efficient and faster than before by only using the information calculation of the data itself. Finally, since each label has a different degree of description for the same instance in real world, to make the description of the instance more accurate, label distribution was used to instead of traditional logical labels. The experimental results show that the proposed algorithm can retain the features with high correlation with the label space, so that the classification accuracy is improved to a certain extent compared with that without feature selection.

Since the ciphertext length of attribute-based encryption scheme from multilinear maps is large, the decryption is inefficient and the scheme has key escrow problem, a key-policy attribute-based encryption scheme from multilinear maps was proposed by using outsourcing technology and user's secret value. The proposed scheme supported general polynomial-size circuit and arbitrary fanout, the private key was generated by key generation center and user. The length of the ciphertext is fixed to | G|+| Z|, compared with the known ciphertext scheme with the minimum ciphertext, the storage cost is decreased by 25% after setting reasonable parameters in accordance with the standards elliptic curves. Users only need to compute transformation ciphertext and the ciphertext is verifiable. The decryption multilinear operation count is only 3, which greatly reduces the computional cost. Selective security is proved in standard model under the multilinear decisional Diffie-Hellman problem. Additionally, it also can be applied in small mobile devices with limited computing capability.

In multi-controller Software Defined Network (SDN), since the existed switch migration strategies always have low efficiency and need to migrate many times which only consider single migration factor, a mechanism of switches migration based on progressive auction named PASMM (Progressive Auction based Switches Migration Mechanism) was proposed. To improve network benefit, the switch migration problem was optimized by auctioning controllers' remaining resources in the mechanism. By increasing the trading price of the over-demanded controllers' resources, PASMM completed the auction and redeployed the controllers and switches. The simulation results show that, compared with some typical switch migration policies, PASMM achieves good load balancing of controllers, reduces the response time of the PACKET_IN messages by an average of 13.5%, and spends the least migration time with the increasing of switches flow requests.

For the inaccurate problem of the estimation of the blood flow velocity which is caused by the clutter signal in ultrasound Color Flow Imaging (CFI), this paper proposed a clutter suppression method based on dynamic region polynomial regression and Singular Value Decomposition (SVD), called ARS algorithm. First, according to the time-domain characteristics and the energy intensity of the echo signal, this method adopted the dynamic partitioning method to distinguish the range of signal; then, according to the divided range, polynomial regression method or SVD method was selected to dynamically reject the clutter signal. This paper made a simulation to compare the proposed method with the projection initialized Infinite Impulse Response (IIR) filter, the non-stationary filter, the regression filter and the SVD algorithm. The experimental results show that the proposed method can completely reject the interference of tissue motion (the velocity is almost zero in the tissue area and the clutter-to-blood ratio is about 5.427 dB after the clutter suppressing is implemented), the estimated maximum blood flow velocity (0.968 m/s) is close to the theoretical value and the blood flow distributes uniformly, the integrity of the blood flow velocity profile can be better maintained and the achieved blood flow velocity map illustrates the authentic flow velocity and high image quality.

A volume measurement algorithm based on edge detection was introduced in this paper. The algorithm was applied to the portable system for measuring bladder volume. In the system, the three-dimensional ultrasound probe performed the scanning with the axial angle of 180 degree. Twelve axial sections of a bladder with the same scanning interval were obtained and used to edge extraction. The group of bladder edges formed a three-dimensional bladder body and the integral method of spatial volume was used to calculate the bladder volume.The average error of the algorithm in clinical diagnosis experiment was within 12%.

The cooperative rollback recovery mechanism based on checkpointing is an effective mechanism for the survivability of distributed system. The existing cooperative rollback recovery mechanism based on checkpointing presumes that the communication channel is reliable. However, this assumption is not always true in actual application scenarios. For the actual application scenarios of distributed system, a reliable assurance model for the survivability of distributed system was proposed, based on the checkpointing-based rollback recovery mechanism. Through the creation of redundant communication channel and process migration mechanism, the proposed model assures the survivability of distributed system in actual application scenarios where the communication channel is not reliable.

To segment moving crowd with different dynamics in complex video surveillance scenes, this paper proposed a crowd motion segmentation algorithm which was based on video particle flow and Finite Time Lyapunov Exponent (FTLE) field. Firstly, video particle flow was used to represent the long-range particle motion estimation. To optimize these particles trajectories, an energy function containing point-based appearance matching and distortion between the particles was minimized. Then the spatial gradient of the particle flow map was solved and the FTLE field was constructed. Finally, the Lagrangian Coherent Structure (LCS) in the FTLE field was used to divide flow into regions of qualitatively different dynamics. The experimental results show that the proposed algorithm can effectively segment crowd flow with different dynamics in complex video surveillance scenes, and it has strong robustness.