The Synthetic Aperture Radar (SAR) image uses Support Vector Machine (SVM) and Markov Random Field (MRF) or Conditional Random Field (CRF) to classify based on feature extraction of coarsely segmented pixel blocks. The traditional method exists the deviation issue of different type pixels inside the same pixel block and it only considers the adjacent area without using global information and structure information. Fully Convolutional Network (FCN) was introduced to solve the deviation problem, and the original classification probability of pixel was gotten by constructing convolutional layers based on pixel level for sample training and using ESAR images as samples. Then CRF-Recurrent Neural Network (CRF-RNN) was introduced as post layer to combine the original classification probability obtained by FCN with full image information transfer and structure information, which was produced by CRF structure. Finally, the RNN iteration was used to further optimize the experimental results. By taking advantages of global information and structure information, the proposed method based on pixel level solved some disadvantages of the traditional classification. The classification accuracy rate of the proposed method was improved by average 6.5 percentage points compared with SVM or CRF. The distance weight of CRF-RNN is fitted by Gaussian kernel, which can not be changed or determined according to the training data, thus it remains some deviation. So a convolutional network based on trainable full image distance weight was proposed to improve CRF-RNN. The experiment results show that the classification accuracy rate of the improved CRF-RNN is further improved by 1.04 percentage points.

The robust optimal control of single Conveyor-Serviced Production Station (CSPS) with uncertain service rate was researched. Under the cases where only the interval of service rate was given and the look-ahead range was controllable, the optimal robust control problem could be described as a mini-max problem by using Semi-Markov Decision Process (SMDP) with uncertain parameters. Global optimization method was adopted to derive the optimal robust control policy when states were dependent. Firstly, the worst performance value was obtained under fixed policy by genetic algorithm. Secondly, according to the obtained worst performance value, the optimal robust control policy was achieved with simulated annealing algorithm. The simulation results show that there is little difference between optimal performance cost of the system whose service rate is fixed as the mean of interval and optimal robust performance cost of the CSPS system with uncertain service rate. Moreover, the difference is getting smaller when the uncertain interval narrows and it means that the global optimization algorithm works effectively.

In the arithmetic for matching multiple patterns based on digital search tree,the physical storage mode of digital search tree is doubly-chained tree. Using the idea of KMP arithmetic, digital search tree has been turned into improved doubly-chained tree through added assistant jump-node. The improved storage mode and arithmetic have quickened the speed of matching, and have implemented non-backtracking in process of matching.