Changzhi Lv

An updating method of self-adaptive background for moving objects detection in video

Background subtraction is a general and simple method in real-time detection of moving object. However, it requires the accurate current background image, and so far, no reasonable approach has been designed and implemented for automatic background updating along with the illumination variance, which limits its applications.To overcome the above problem, a new self-adaptive background approximating and updating algorithm based on static background and kalman self-adaptive background (KAB) is presented in this paper. Moreover, the two renewal rates in KAB are obtained by the cumulants of background subtraction in object region and background region. Experimental results demonstrate that the proposed new background updating method can update the background exactly and quickly along with the variance of illumination, the renewal rates can vary with the video automatically and they bring certain noise immunity to the new dynamic background.

Wall-adherent cells segmentation based on cross-entropy and watershed transform

The microscopic image processing technology is a new solving approach to segment and count wall-adherent cells in anti-virus experiment in vitro. But the segmentation is very stubborn because of the cellspsila multiformity.This paper presents a segmentation strategy based on cross-entropy and watershed transform to segment and count the wall-adherent cells. Firstly, top-hat transform is used to enhance the original cells microscopic image. Suppose the conditional distributions of object and background are modeled with normal distributions, maximum between-class cross-entropy threshold segments the image into binary one. Then morphological filters reduce the burrs and holes in binary image and watershed transform further segments the cells by single-pixel wide edges. Finally, the cells are counted by labeling them.The experiments show that this strategy is effective, simply and strongly adaptive to lighting. The segmentation boundaries are continuious and the cellspsila shapes are well kept.

A Wireless Collector for EEG Signal of Freely Behaving White Mice

Study EEG signal is an approach to understand the relationship between neurons electron activities and animals physiological and psychological state. A wireless EEG signal collector is designed in this paper. The collector contains microelectrodes set, front circuit, low-pass filter circuit, data collecting and transmitting circuit, and power supply circuit. When the two microelectrodes are pushed by propeller and inserted into the brain of white mice and the EEG signal is detected. It is amplified more than 10 thousands times by two level amplifiers and filtered by a HP filter and a LP filter to remove DC and high frequency interfere. Then the strong EEG signal is converted to digtal data and transmitted to host computer.The collector is wireless and miniature and the animal can behave freely. It has been used in our EEG signal acquisition system and shown good performance in carrying and communicating. It also can be used to acquire other animals EEG signal.

Locating the Ultrasonic Echo Head Wave of Bottom Sediment in Drilling Hole Based on Local SNR

Locating the head wave of ultrasonic echo accurately is the basis of measuring the thickness of sediment in drilling hole precisely. This paper presents a method to locate the echo head wave by local signal-to-noise ratio (L-SNR). The original signal is filtered by a parameters constrained Gabor filter and transformed into frequency domain by windowed Fourier transform firstly, and then the L-SNR is computed. The peaks in L-SNR function just represent and locate the beginnings of echos. The experiments results show that the L-SNR has sharp peaks at the beginnings of the echos indeed and the locating based on these peaks is high precise. This paper has also indicated that the length of the time window should be the 4~8 times of the reciprocal of transducer bandwidth.

Wall-Adherent Cells Segmentation Based on BP Neural Network

Anti-virus experiment in vitro is a common way to screen and identify antiviral drugs. Most of the cells in the experiments are wall-adherent. Segmenting, recognizing and counting this wall-adherent cells in micrograph successfully can cut down the time and cost of the experiment. It is much difficult to segment this kind of image because of the wall-adherent cells characteristics. In this paper, some works have been done to use BP neural network to segment wall-adherent cells. Firstly, three layers BP network has been designed and features are extracted for segmentation. The weights and thresholds of the network are determined by training it with hundreds of samples. Other three typical wall-adherent cell images are respectively input into the designed BP network to test its segmentation performance. In the segmentation test result images, the cells are well segmented not only in clear region, but also in polluted region and small fracted region. The test results show that the designed BP network is much effective for wall- adherent cell image segmentation.

A design of geophone tester based on FPGA

Geophone is a specific sensor in geophysical exploration. The geophone tester is mainly used to test the performance parameters of geophone such as DC resistance, natural frequency, damping coefficient, sensitivity, distortion and so on. Some domestic testers are big, low precision and high power consumption and not suitable for outdoors. The testers abroad are expensive and their performances are not stable. This paper designs a high performance-cost ratio tester based on FPGA. It adopts EDA technology and programmable Cyclone II series FPGA on the platforms of DSP_Builder and Quartus II. On the FPGA, there integrates 24-bit data collector, FIR filter, 32 points FFT, UART and etc. Some testing experiments are fulfilled and the results show that this portable tester is of high precision, low power, and it is easy to be upgraded and maintained.

Fast registration of UAV aerial images based on improved optical-flow model combined with feature-point matching

With a large number of registration algorithms proposed, image registration techniques have achieved rapid development. However, there still exist many deficiencies in aerial images registration where high speed and accuracy are difficult to simultaneously achieve for real-time processing. In order to achieve large-scale and high-precision image registration for unmanned aerial vehicle(UAV) aerial images, a novel and fast sub-pixel image registration algorithm based on improved optical-flow model combined with feature-point matching is proposed in this paper. Firstly, the coarse selection at the feature level is achieved by using the feature-point model, which reduces the number of non-feature points so as to speed up the coarse registration process. Then, the improved pyramid optical-flow model is adopted in the neighborhood of the coarse point, and the sub-pixel fast location is achieved by the bidirectional search strategy. Simulation experiment results show that compared with common image registration based LK optical-flow or feature-point matching, our proposed algorithm will greatly reduce space complexity and time complexity without losing accuracy.

Intercollicular nucleus electric stimulation encoded “walk forward” commands in pigeons

The bio-robot research field is growing. Robo-pigeons have been successfully programmed to turn left or right; however, a satisfactory method of commanding a robo-pigeon to walk forward is still lacking. This problem has become a roadblock to progress in bio-robot research and applications. In mammals, the midbrain periaqueductal gray region (PAG) plays a key role in mediating defensive reactions in response to fear and anxiety. The avian intercollicular nucleus (ICo) is thought to correspond to the PAG. In this study, we found that microstimulating the ICo could successfully induce a robo-pigeon to walk forward. Compared with stimulation of the previously used archistriatum, the response time was considerably shorter and the behavior accuracy significantly higher. This paper describes in detail the process of controlling a robo-pigeon such that it walks forward and backward along a prescribed straight line. From the results, we draw the conclusion that the ICo is suitable for prompting the “walk forward” order in robo-pigeons.

SVD Filter Based on Noise Singular Values Clustering

Aiming at determining the optical reconstruction rank of singular value matrix in SVD filter, a new method based on noise singular values clustering is proposed in this paper. Variable forward standard deviation (F-Std) has been defined and the mutation point in F-Std is used to separate signal singular values from noise singular values. Experiments have been simulated by some noisy signals using the new SVD filter and other two commonly used methods, the results and root mean square errors of these filters are shown that the new SVD filter is not only optimal or near to optimal and more effective than the two methods but also robust to the shape of the trajectory matrix even when the signal’s SNR varies from very high to -11.7dB.

A distortion tester of geophone based on FPGA

Distortion is an important performance of geophone and it can effect the accuracy of the data obtaining in seismic exploration directly. In this paper, a distortion tester based on FPGA (Field Programmable Gate Array) has been designed. On the base of the dynamic and equivalent circuit models of geophone and the analysis of windows, the measurement is exciting the geophone by a sine current signal, transforming the windowed response sequence by FFT (Fast Fourier Transform) and calculating the distortion of response. The DDS(Direct Digital Synthesis) to generate excitation signal and FFT are realized on the FPGA. Experiments to test the distortion of GP903/4A-10Hz geophone have been done and the results show that the measurement is sample and test precision is high. The application of FPGA also makes the tester low cost, low power, small size and it is easy to be upgraded.