Search

Agricultural development and food security in karst areas affect the well-being of local people. In order to clarify the spatial correlation between soil physical-chemical indicators, samples of soil were taken to analyze the soil pH, soil electric conductivity, soil temperature, and soil moisture volume content at different soil depths from 0–30 cm in Guohua Karst Demonstration Zone as research subjects. Combining indoor sampling point distribution and field monitoring, we comprehensively used classical statistics, spatial auto-regression analysis method, and “3S” technology to establish the spatial auto-regression models of soil physical-chemical indicators of different soil depths in a micro-scale format. The results showed that the mean value of soil pH and soil temperature had a ow variations, while the coefficients of variation decreased with the increase of soil depth. Soil moisture volume content showed medium variation to intense variation, and soil electric conductivity had intense variations. The soil physical-chemical indicators ranging from 5–10 cm soil depths were both in accordance with the spatial lag auto-regression models. The soil physical-chemical indicators from 20–30 cm soil depths were both in accordance with the spatial residual autoregression models.

In this paper, an adaptive dynamic surface control scheme is proposed for a class of output feedback systems with input unmodeled dynamics. By constructing the state observer, unmeasurable states are estimated. By introducing the first order filter in traditional backstepping design, the repeated differentiations of virtual control is avoided. Using dynamic surface control, the disturbances induced by input unmodeled dynamics and unknown continuous function are effectively dealt with. By theoretical analysis, it is shown that all the signals in the closed-loop system are semi-global uniformly ultimately bounded.

Directing towards the characteristics and importance of copper strips surface inspection, this paper proposes a copper strips surface inspection system based on artificial intelligence from the standpoint of artificial intelligence. It uses computer vision to capture defects images, uses Hu invariant moments as the eigenvectors and uses BP neural network based on genetic algorithm combined with expert system to train and learn the samples. The trained weights are used to diagnose as the knowledge base and a copper surface defects inspection system based on intelligent fusion is constructed . The experimental results proved that intelligent fusion model can compliment all kinds of intelligent models and has a better performance than single model on copper strips surface inspection.

In this paper, novel video transmission architecture is proposed to meet the error robust requirement of wireless video communications. This architecture uses user-defined protocol to provide efficient bandwidth access from wireless link. By taking the time-varying wireless channel condition and video codec characteristics into account, first, we present an end-to-end transport architecture for video streaming over wireless network. Second, we adopt channel of Bernoulli channel model from application layers of 3G networks. And we propose a dynamically estimate the time-varying wireless network condition through estimate of throughout, RTT and merit of packet delay time. Third, we use H.264 as an experiment platform. A data partition scheme is also used to enhance error resilience performance. This architecture is also suitable for various motion compensation based standards like H.261, H.263 series, MPEG-1, MPEG-2, MPEG-4. The simulation results show our strategies can significantly improve the robustness of wireless video transmission.

Recently, it becomes more important to ensure the quality of the products as copper strip manufacturing has been highly developed. The most difficult problem in process control and automatic inspection is classification of surface defects, so we develop an improved RBF (radial basis function) neural network classifier based on SVM (support vector machine) to automatically learn complicated defect patterns and use pseudo Zernike moment invariant as the defect feature. The optimal initial parameters of RBF network are gained through SVM, which has resolved the problems in traditional methods, e.g. long learning time, and easily getting into local minimum, etc. Furthermore, a BP learning algorithm is presented to adjust these hidden node parameters as well as the weights of the SVM-RBF. The experimental results show that the method is effective.