Group decision making of domain experts is the most direct approach to determine loss function in three-way decision problems. Different from linguistic variable model and fuzzy set model with single uncertainty, expert evaluations described by cloud model can reflect the complex uncertainty form in cognitive process, and the synthetic evaluation function can be obtained by cloud aggregation. However, numerical characteristics only are performed simple linear combination in current cloud aggregation methods, leading the lack of the description of concept semantic differences and the difficulty to obtain convincing results. Therefore firstly, the weighted distance sum was proved to be a convex function in the distance space of cloud model. And the aggregational cloud model was defined as the minimum point of that function. Then, this definition was generalized to the multi-cloud model scenario, and a cloud aggregation method namely density center based cloud aggregation method was proposed. In group decision making, the proposed method obtains the most accurate synthetic evaluations with the highest similarity between synthetic evaluation and basic evaluation, providing a novel semantic interpretation of the determination of loss function. The experimental results show that the misclassification rate of the three-way decision with loss function determined by the proposed method is the lowest compared with simple linear combination and rational granularity methods.

Aiming at the problem of cryptographic performance being affected by the excessive interrupt transmission cost of the cipher device in virtual environment, an interrupt path optimization method for virtual cryptographic device with Reducing Context Switching (RCS) was proposed. Firstly, a host to Virtual Cipher Machine (VCM) relationship mapping table was established in the kernel of the virtual machine. Then, the types of the interrupt requests that the host transmits to the VCM were judged by the relational mapping table, and the unassigned types in VCM were registered. Finally, the interrupts were handled by the VCM interrupt handler directly. In the process, the system context switching overhead was reduced due to the host intervening and the cryptographic performance was improved. The speed at which the VCM executes the encryption was selected as a performance reference in the experiment. The results show that the speed of VCM using Advanced Encryption Standard (AES) algorithm is increased by 16.35% and that using Secure Hash Algorithm (SHA256) is increased by 12.25%.

Focusing on the issue of target motion feature extraction of video sequences in complex scene, and referring to the motion perception of biological vision system to the moving video targets, the traditional primary Visual cortex (V1) cell model of visual cortex was improved and a novel method of random-dot motion feature extraction based on the mechanism of biological visual cortex was proposed. Firstly, the spatial-temporal filter and half-squaring operation combined with normalization were adopted to simulate the linearity and nonlinearity of neuron's receptive field. Then, a universal V1 cell model was obtained by adding a directional selectivity adjustable parameter to the output weight, which solved the problem of the single direction selectivity and the disability to respond correctly to multi-direction motion in the traditional model. The simulation results show that the analog outputs of proposed model are almost consistent with the experimental data of biology, which indicates that the proposed model can simulate the V1 neurons of different direction selectivity and extract motion features well from random-dot video sequences with complex motion morphs. The proposed method can provide new idea for processing feature information of optical flow, extract motion feature of video sequence and track its object effectively.

Aiming at improving the robustness in pre-processing and extracting features sufficiently for Synthetic Aperture Radar (SAR) images, an automatic target recognition algorithm for SAR images based on Deep Belief Network (DBN) was proposed. Firstly, a non-local means image despeckling algorithm was proposed based on Dual-Tree Complex Wavelet Transformation (DT-CWT); then combined with the estimation of the object azimuth, a robust process on original data was achieved; finally a multi-layer DBN was applied to extract the deeply abstract visual information as features to complete target recognition. The experiments were conducted on three Moving and Stationary Target Acquisition and Recognition (MSTAR) databases. The results show that the algorithm performs efficiently with high accuracy and robustness.