Chen Tian-hua

License Plate Recognition Based on Edge Detection Algorithm

With the development of vehicles and the increasing number of cars in modern society, people pay more and more attention to the vehicle license plate recognition system. Vehicle license plate recognition is divided into three parts: license positioning, character segmentation and character recognition. This paper proposes a new way of vehicle license plate recognition. In license positioning, the license positioning method is based on median filtering double edge detection, in character segmentation, it adopts a combinative method of locating the original level and improved vertical projection segmentation algorithm, in character recognition, the recognition method is based on classification template matching of font characteristics. The proposed license recognition method focuses on the timeliness of recognition process. Its smaller calculation largely reduces the time of license recognition, with high accuracy and precision rate at the same time.

Image Fusion Method Based on NSCT and Robustness Analysis

Infrared and visible image fusion technology can effectively improve the image contrast and clarity, and enhance the night vision effective. Non-sub sampled contourlet transform (NSCT) in image fusion field has made some achievements. A regional standard deviation-weighted image fusion method based on non-sub sampled Contourlet transform was proposed, and the robustness of the method was analyzed. First, the reiterated infrared and visible images from the same scene were transformed by non-sub sampled contourlet transform, followed the approximate weight averaged, high-frequency detail components in accordance with the weighted of the regional standard deviation proportion, then the fusion image is obtained by inverse non-sub sampled contourlet transform, Finally, the fusion images were compared with the results obtained by Laplace transform, wavelet transform and contourlet transform through a large number of experiments, and the robustness analysis was done through the noise test. The results show that: non-sub sampled contourlet transform can achieve better fusion effect, and high robustness.

The Sorting Method of ECG Signals Based on Neural Network

This paper gives a new way of recognizing and sorting of ECG signals, using ECG signals analysis and improved back propagation artificial neural network. Firstly, the ECG signals has to be digitally filtered to eliminating the 50 Hz noise. The neural network adopted is improved back propagation three-layer artificial neural network, with an optimized BPN4-3-2 network topology. Get the four characteristic parameters: span of QRS wave, interval of R-R segment, voltage of S-T segment, slop of S-T from filtered ECG signals to act as the input parameters of the input layer. After about 1200 times training of 16 samples, the network can correctly distinguishes coronary heart disease sample from non-coronary heart disease samples. Meanwhile, this network can also be applied to sort the 10 non-trained ECG signals samples correctly. The results of experiment show, that the method of recognizing and sorting ECG signals discussed in this paper totally accords with the clinic diagnosis. So it is a new, promising, non-invasive method of diagnosing coronary heart disease.

Image fusion based on regional energy and standard deviation

Image fusion technology of infrared and visible light could enhance the drivers visual effects in low visibility circumstance, and reduce the probability of the traffic accidents. Focused on the application of wavelet transform in image fusion, a image fusion method was proposed based on a regional energy and standard deviation for the infrared and visual light images. In this method, low-frequency approximation component of wavelet transform were obtained by a regional energy-weighted fusion rule, and three high-frequency details components were composed by utilizing the regional standard deviation-weighted fusion rule. Through the two groups of infrared and visual light image fusion experiments show that: the standard deviation of the image produced by the method increased 41.8% and 67.2% and the average brightness increased from 34.6% to 55.1% compared with simple method, respectively.

DSP-based variable structure control of switched reluctance motor

This paper analyzes highlight the shortcomings of the nonlinear and variable parameters for switched reluctance motor(SRM), and the advantages of variable structure control systems, presents a control method for SRM based on DSP and variable structure control method, Control system using TMS320F2407 chip, full use of the DSP chip resources and the advantages of sliding mode variable structure control system, the hardware structure is simple, and achieve the speed control of SRM. Experiments show that the variable structure control method has better control of SRM dynamic performance and steady precision, and parameter perturbations and external disturbances has strong robustness.

Notice of Retraction The Intellectual Detection and Classification Based on Neural Network for Freshness of Pork

In this paper, CCD photoelectric detection techniques that can be characterized by a number of characteristics of fresh pork in the amount of testing, using digital image processing techniques to detect the characteristics of the amount of processing, the final study of using neural network technology micro-parameters of non-coherent multi-data fusion detection methods, in order to achieve the freshness of the meat inspection classification identification.

Image Fusion Technology Based on Bio-inspired Features

Infrared and visible image fusion, as a powerful tool for the object detection and recognition, has developed with the advent of various imaging modalities. However, resulting images of traditional methods are always difficult to compromise between multimodalities. This paper has solved this problem by a variable-weight fusion rule based on the non-sub sampled contourlet transform (NSCT). The original images are combined in the multiscaled space and the fused image is obtained in the bio-inspired feature frame. Validation experiments on infrared and visible images are for two purposes: the comparison among different fusion rules and the impact of the multiscaled analysis in infrared and visible image fusion. In order to evaluate the proposed method, information entropy (IE), standard deviation (STD), spatial frequency (SF) and mutual information (MI) are adopted to compare with Laplace, wavelet, and NSCT et al. Results are shown that all evaluation value of the proposed method is higher than that of other methods, and it is a better image fusion method.

Study on the Robustness of Hydraulic Turntable Variable Structure Control

This paper establishes the mathematical model of the three degrees freedom hydraulic turntable system, and presents a new design method and procedures of the variable structure control(VSC) system, and a new control law based Lyapunov theory. In this paper, simulation experiments of three degree freedom electrical-hydraulic turntables are conducted, and researches the robustness characteristic of variable structure control in theory. Research and experiments have shown that the variable structure control system is distinguished by its excellent dynamic and steady-state characteristics, and is insensitivity to variation in the plant parameters and to external disturbances.

Detection of Meat Fresh Degree Based on Neural Network

In order to find a quick valid scientific way to identify meat freshness, the article analyses the measuring mechanism of identifying to fresh degree of meat, and designs one set of intellectual detection and identification system that is based on electronic information technology, photoelectric detection technique, image processing technology and neural network model recognition technology. Through the test and analysis of the pork sample, it proves that this system is real-time in discerning the fresh degree of meat accurately. In addition, this method can apply to other relevant fields. The research result has very important reality and long-term meaning.