Focused on the existing problems that there are few image datasets of citrus diseases and insect pests, the targets of diseases and pests are complex and scattered, and are difficult to realize automatic location and segmentation, a segmentation method of agricultural citrus disease and pest areas based on Superpixel Fast Fuzzy C-means Clustering (SFFCM) and Support Vector Machine (SVM) was proposed. This method made full use of the advantages of SFFCM algorithm, which was fast and robust, and integrated the characteristics of spatial information, meanwhile, it did not require manual selection of samples in image segmentation like the traditional SVM. Firstly, the improved SFFCM segmentation algorithm was used to pre-segment the image to be segmented to obtain the foreground and background regions. Then, the erosion and dilation operations in morphology were used to narrow these two areas, and the training samples were automatically selected for SVM model training. Finally, the trained SVM classifier was used to segment the entire image. Experimental results show that compared with the following three methods:Fast and Robust Fuzzy C-means Clustering (FRFCM), the original SFFCM and Edge Guidance Network (EGNet), the proposed method has the average recall of 0.937 1, average precision of 0.941 8 and the average accuracy of 0.930 3, all of which are better than those of the comparison methods.

For the increasingly severe parking problem, a method of parking lot space detection based on a modified convolutional neural network was proposed. Firstly, based on the characteristic that a parking lot only needs to be denoted by two states, a concept of Mini Convolutional Neural Network (MCNN) was proposed by improving the traditional CNN. Secondly, the number of network parameters was decreased to reduce the training and recognition time, a local response normalization layer was added to the network to enhance brightness correction, and the small convolution kernel was utilized to get more details of the image. Finally, the video frame was manually masked and cut into separate parking lots by edge detection. Then the trained MCNN was used for parking lot recognition. Experimental results show that the proposed method can improve the recognition rate by 3-8 percentage points compared with the traditional machine learning methods, and the network parameters of MCNN is only 1/1000 of the conventionally used convolutional model. In several different environments discussed in this paper, the recognition rate maintains above 92%. The experimental result shows that the MCNN can be transplanted to a low-configuration camera to achieve automatic parking space detection.

In order to reduce the redundant connections and unnecessary computing cost, quantum-inspired clonal algorithm was applied to optimize neural networks. By generating neural network weights which have certain sparse ratio, the algorithm not only effectively removed redundant neural network connections and hidden layer nodes, but also improved the learning efficiency of neural network, the approximation of function accuracy and generalization ability. This method had been applied to wild relics security system of Emperor Qinshihuang's mausoleum site museum, and the results show that the method can raise the probability of target classification and reduce the false alarm rate.

Images usually have multiple connected regions of the same color. For the problem that Pulse Coupled Neural Networks (PCNN) cannot abstract these areas separately, a PCNN model with improved traversal process was proposed. By introduceing the depth-first search traversal algorithm, multi-unconnected regions were activated on different layers, so as to achieve a separation. Finally, the new model was improved again for the effect of image noise. The activated scope in each layer was used to detect noisy pixels, and then the mean-shift algorithm was introduced to eliminate the noisy pixels. The separation effect of multi-regions with the same color in the image and the ability to eliminate noise has been verified by experiment.

Scheduling of Imaging Satellite (IS) involves many complex constraints. In the battle fields, it is difficult to schedule imaging satellites to satisfy the requirement of strategic decision-making. In view of this problem, an algorithm based on an improved ant colony algorithm with elitist strategy, which focused on the scheduling problem of multiple imaging satellites, was proposed. A specific description on the algorithm's state transformation rules and pheromone update rules was given. A disposal flow of task roadmap based on heuristic method was proposed to generate scheduling plan and evaluate the roadmap, and the result of which fed back to the path search phase. A case was given to compare the proposed algorithm with greedy algorithm and genetic algorithm to validate that this one can help acquire results of high quality.

The authentication and key agreement protocol adopted by 3rd Generation Partnership Project (3GPP) System Architecture Evolution (SAE) Release 8 standard was analyzed in contrast with 3G, and several security defects in SAE protocol were pointed out, then an improved 3GPP SAE authentication and key agreement protocol was put forward based on public key cryptosystem. In the new protocol, user’s identity information and authentication vector in network domain were encrypted based on public key cryptosystem, public parent key adopted in local authentication was generated by random data. The security and efficiency of the proposed new scheme was analyzed at last. The analysis results show that the proposal can effectively solve the problems mentioned above and improve the security of protocol with less cost.