The traditional vision algorithms have low accuracy and are greatly affected by environmental factors during the detection of long and slender power lines in complex scenes, and the existing power line detection algorithms based on deep learning are not efficient. In order to solve the problems, an end-to-end fully convolutional neural network model was proposed which was suitable for power line detection on mobile terminals. Firstly, a symmetrical encoder-decoder structure was adopted. In the encoder part, the max-pooling layer was used for down-sampling, so as to extract multi-scale features. In the decoder part, the max-pooling indices based non-linear up-sampling was used to fuse multi-scale features layer by layer to restore the image details. Then, a weighted loss function was adopted to train the model, thereby solving the imbalance problem between power line pixels and background pixels. Finally, a power line dataset with complex background and pixel-level labels was constructed to train and evaluate the model, and a public power line dataset was relabeled as a different source test set. Compared with a model named Dilated ConvNet for power line semantic segmentation on mobile devices, the proposed model has the prediction speed for 512×512 resolution images on the mobile device GPU NVIDIA JetsonTX2 twice that of Dilated ConvNet, which is 8.2 frame/s; the proposed model achieves a mean Intersection over Union (mIoU) of 0.857 3, F1 score of 0.844 7, Average Precision (AP) of 0.927 9 on the same source test set, which are increased by 0.011, 0.014 and 0.008 respectively; and the proposed model achieves mIoU of 0.724 4, F1 score of 0.634 1, AP of 0.664 4 on the public test set, which are increased by 0.004, 0.007 and 0.032 respectively. Experimental results show that the proposed model has better performance of real-time power line segmentation on mobile terminals.

A strategy for integrated berth and quay-crane scheduling was proposed to cope with unreasonable allocation of port resources in container terminals. First, a nonlinear mixed integer programming model which aims at minimizing the port operational cost was presented. And the loading and unloading cost of quay-crane was considered in the objective of our model. To make the model more realistic, the handling time of a vessel was assumed to depend on the number of assigned quay-cranes. Second, an improved genetic algorithm based on extenics dependent function was used to solve this model. In this algorithm, infeasible solutions play an important role. They were evaluated by their extenics dependent degrees. Some infeasible solutions were always contained in the population to maintain the diversity of the population. This improved local search ability of traditional genetic algorithm. At last, the effectiveness and efficiency of the proposed model and algorithm were testified by several test instances. Compared with the model without considering the loading and unloading cost of quay-crane, the waste of resource is effectively reduced.

In order to enhance the performance of automobile suspension, a glowworm-PID (Proportion-Integral-Differentiation) algorithm was put forward. Firstly, on the basis of analyzing the basic principle of glowworm-PID algorithm, the glowworm-PID algorithm for motor actuator suspension was developed and the steps and flow diagram of the algorithm were given. Secondly, traditional motor actuator was improved and mathematical model and simulation model of 4 degrees of freedom motor actuator suspension were built. Hardware-In-the-Loop Simulation (HILS) algorithm testing system of active suspension was developed by taking dSPACE as the carrier of the simulation model. Lastly, simulation experiments of testing Glowworm-PID algorithm were carried out on the testing system with different vehicle data and road input. The simulation results indicated that Glowworm-PID algorithm designed in this paper could reduce acceleration of vehicle bodies, working space of suspensions and moving displacement of tires, and so on.

Basic Input Output System (BIOS) trapdoor has huge impact on computer system, and it is difficult to detect the existence of BIOS trapdoor effectively with the existing tools. After researching BIOS structure and BIOS code obfuscation technique based on reverse analysis, BIOS trapdoors were divided into module-level BIOS trapdoor and instruction-level BIOS trapdoor according to implementation granularity, followed by analyzing the implementation principle and characteristics of these two BIOS trapdoors in detail. Finally the detection method of module-level trapdoor based on analyzing module structure and the detection method of instruction-level trapdoor based on integrity measurement were presented. The experimental results show that these two methods can detect the existence of their corresponding BIOS trapdoors effectively.

According to the blind extracting issues within the spread-spectrum watermark, a kind of blind extracting algorithm which could be used in the extraction of the digital audio signals was proposed. In the algorithm, wavelet transform was applied to the audio document, then the Discrete Cosine Transform (DCT) was used to its low-frequency coefficient. Afterwards, the fifth coefficient was got and it was used to hide the watermark information being spectrum spread. As the spread-spectrum sequence and its length were unknown during the extraction, spectrum-reprocessing and Singular Value Decomposition (SVD) were introduced to estimate the spread-spectrum using in the embedding process, and the blind extraction to the spread-spectrum watermark of the given digital signal was fulfilled. The simulation results show that with unknown spread-spectrum parameter, watermark image with Normalized Coefficient (NC) of one can be extracted, and it is of strong robustness. Under the attacks of noises and low-pass filter, the accuracy rate of the estimating spread-spectrum sequence is over 90%, which guarantees the recovery of clear water mark image with normalization coefficient higher than 0.98.