Search

Edge detection is one of the most central branches in the field of image processing. The traditional edge detection algorithm is studied for its poor resistance to pretzel noise and poor adaptability of threshold selection. An improved image edge detection algorithm based on Canny is proposed, which firstly combines improved adaptive median filtering to reduce image noise. Then, the local adaptive thresholds are used to detect the edges of the image. The comparison of experimental results shows that the improved algorithm, which addresses the shortcomings of the traditional Canny algorithm, is better at removing the interference of pepper noise and preserving the image edge details.

The rapid growth of emerging technologies with artificial intelligence and big data as new representatives has brought great influence to all walks of life, as well as the education. College sports is an important part of the education system and deeply affected by the artificial intelligence. The paper will briefly summarize the current status of college physical education in China and discuss how to innovate its methods in the era of artificial intelligence.

Bridges have become an important part of urban transport, and the length and number of Chinese bridges are world leaders. In the event of an accident, it will cause significant losses. Therefore, it is especially important to prevent the health of the bridge and to repair it in advance. This paper describes the system monitoring of bridge monitoring based on situational awareness-based big data. Based on the development process of bridge monitoring, a big data bridge processing system based on situational awareness technology is proposed. The system uses the situational awareness architecture to take machine learning as the core, and the data collected by the sensor is imported into the bridge health monitoring system. The monitoring data is processed by the expert system of machine learning, and the processing suggestions are given in the human-computer interaction interface. Finally, the paper gives the problems to be solved and future prospects of bridge monitoring.

Multivariate statistical analysis is often used in fault diagnosis. PCA is a commonly used multivariate statistical analysis method, based on the threshold method for fault diagnosis. But the PCA is the original space of the equivalent that didn't increase any amount of information, often resulting in a lot of minor failures difficult to diagnose. Deep Learning has a successful application in pattern recognition, linear representation of detail based on multi-level network. But Deep Learning is not explainable, only to give training results and there is no theoretical basis. On the basis of the original PCA, firstly expands and then reduces the dimension. The former is intend to fully express the information which cannot be expressed in original space by adding a projection, two projection,..., on the basis of original projection, and so on. And then generate a multi-level projection frame representing the space from the sample space high-dimensional feature. Experiment shows that the minor faults that cannot be detected in the original projection space can be detected under the projection frame of the new multi-level high-dimensional feature, and the detection rate of fault detection is improved.

In this paper, from the problem of the orientation of sudden pollution source which continuously release pollutants in three-dimensional space, while the orientation of the pollution source in the real environment is random and complex under the influence of the internal movement factors of the lake reservoir fluid, the problems are discussed from two aspects. Firstly, the nonlinear least squares estimation is carried out based on the pollution source diffusion model. Then, the sequential idea is combined with the sequential unscented Kalman filter, and the pollution source is precisely orientated by using the sequential unscented Kalman filter. Secondly, under the condition that the pollutant diffusion model is unknown, the diffusion model of pollutants in the unknown diffusion model is trained by ELM to solve the problem of the orientation of sudden pollution sources. The simulation results show that the ELM trains a pollutant concentration diffusion equation, which can effectively avoid the influence of internal movement, wind or other factors, and more in line with the actual environment. After training the concentration diffusion model, it is possible to reduce the complexity of numerical calculation, improve the performance of orientation and improve the robustness of orientating based on the application of the sequential unscented Kalman filter.

Owing to the complexity of sampling processes and running environment of the system, it is hard to avoid abnormal data such as outliers as well as patchy outliers, which widely emerge out of the sampling series. These abnormal data have remarkably bad impact on state estimation, process monitoring, process control as well as safe running of the dynamic system. In this paper, a detailed review is given to describe some research courses and prominent approaches and status quo of outlier-tolerant computation, such as outlier-tolerant estimation, outlier-tolerant identification, outlier-tolerant filtering, outlier-tolerant prediction, etc. At the end of this paper, three open problems are pointed out.

This paper presents a rapidly and lower neural networks to treat those waste water index that is difficult to be measured. Model called soft sensor is composited two parts: one is used to estimate the principal linear output, the other one is used to adjust estimated error to obtain better accuracy. Selection of features that effects greatly computation scale and predict accuracy is discussed also to achieve a better result. Finally, an experiment is provided for testing. Simulation results show the method proposed here is valid and has good performance.

A power converter is a time-varying system which featured with typical characteristic of nonlinearity. Therefore, it lacks of a systematic way to investigating the controllability due to the limitations of linear circuit theory. In order to study the controllability of three-phase inverters in this paper, the three-phase three-wire inverter and the three-phase four-wire inverter are modeled as switched linear systems. Then, the state controllability of the two inverters are studied based on the controllability criterion derived from the switched system theory. Thereafter, the controllability canonical decomposition is conducted when the system is not completely state-controllable. The simulation results verified the correct of the analysis.

In this paper, the problem of stabilizing two agents with a beacon into a triangular formation is studied, in which the beacon is static and the specified triangle is characterized only by interior angles constraints. Assuming that each agent can measure the bearing of the beacon in its local frame. Meanwhile, the two agents neighbor each other, that is, they communicate with each other so that they can measure the bearing angle respectively and exchange the measured angle information each other. A control strategy is proposed utilizing only angle information which steers the two agents to form a triangular formation with the beacon. Both theoretical analysis and simulations are provided to demonstrate the effectiveness of the control algorithm.

This paper presents a biologically plausible model for the structural information processing in early vision. Our investigation on the frequency spectrum of natural images filtered by the retina shows that the DC component containing much redundancy information and the high frequency components containing much noisy information are reduced, while the middle and low frequency components containing much structural information are enhanced. Simple cells in the primary visual cortex (V1) extract structural primitives from the filtered signals resulting in the emergence of diverse receptive field shapes. We name these structural primitives as structors, and study the neural mechanisms responsible for this diversity of V1 simple cell receptive field shapes. Sparse coding with the L0-norm constraint is reexamined which suggests that the local structure of natural images is determined by few structors regardless of their coefficients. We perform an analysis on the spatial distribution of the input signal and prove that signals in the neighborhood of a special structor has a star shape and peaks at the structor. That is, the structors are the modes of the probability density function of the input signal, and learning the structors can be interpreted as mode detection. Mean sift method is applied to detect modes, and the updating rule for the mean shift appears to be Hebbian. We propose the Hebbian-based mean shift to simulate the emergence of the diversity of simple cell receptive field shapes. The simulation results demonstrate the robustness of the proposed algorithm in producing both Gabor-like and blob-like structors.

Alkaline solution of a rare earth (RE) and some northern mixed and heating method of rare-earth ore are used in preparation for a new type of alkaline RE coagulant. Utilize a single factor experiment and orthogonal experiment to decide the best of preparation conditions. Take the coking wastewater for example, comparision with the usual coagulant PFS and PAFC in coagulation test. Experiments show that alkaline earth coagulant coagulation is obvious better than polymerization ferrous sulfate (PFS) and polymerization ferrum aluminum coagulant (PFAC) and has high turbidity removal efficiency, fast flocs formation speed, good performance and other settlement characteristics.