Moving Target Detection Method Research Articles

Research on Video Detection Method of Moving Target Oriented to Substation

With the promotion of unattended substations across the country, intrusion detection technology for people in important areas of substations has become an important research area. This article focuses on the problem of personnel entering important areas of substations for the purpose of man-made destruction, and proposes an improved two-frame difference method A moving target detection method combined with the background subtraction method. This method is based on the two-frame difference method, taking into account the impact of changes in illumination, and avoids the image sequence by combining the difference map of the background subtraction method and the two-frame difference method. The “double shadow” phenomenon caused by the difference method can also avoid the situation that the background subtraction method misjudges the sudden change of light as a moving target, improves the recognition accuracy, and can meet the real-time requirements of the substation intelligent monitoring system.

Deep Learning-Driven Gaussian Modeling and Improved Motion Detection Algorithm of the Three-Frame Difference Method

To enhance the effect of motion detection, a Gaussian modeling algorithm is proposed to fix holes and breaks caused by the conventional frame difference method. The proposed algorithm uses an improved three-frame difference method. A three-frame image sequence with one frame interval is selected for pairwise difference calculation. The logical “OR” operation is used to achieve fast motion detection and to reduce voids and fractures. The Gaussian algorithm establishes an adaptive learning model to make the size and contour of the motion detection more accurate. The motion extracted by the improved three-frame difference method and Gaussian model is logically summed to obtain the final motion foreground picture. Moreover, a moving target detection method, based on the U-Net deep learning network, is proposed to reduce the dependency of deep learning on the number of training datasets. It helps the algorithm to train models on small datasets. Next, it calculates the ratio of the number of positive and negative samples in the dataset and uses the reciprocal of the ratio as the sample weight to deal with the imbalance of positive and negative samples. Finally, a threshold is set to predict the results for obtaining the moving object detection accuracy. Experimental results show that the algorithm can suppress the generation and rupture of holes and reduce the noise. Also, it can quickly and accurately detect movement to meet the design requirements.

Using Sensor Network in Motion Detection Based on Deep Full Convolutional Network Model

Aiming at the shortcomings of traditional moving target detection methods in complex scenes such as low detection accuracy and high complexity, and not considering the overall structure information of the video frame image, this paper proposes a moving-target detection based on sensor network. First, a low-power motion detection wireless sensor network node is designed to obtain motion detection information in real time. Secondly, the background of the video scene is quickly extracted by the time domain averaging method, and the video sequence and the background image are channel-merged to construct a deep full convolutional network model. Finally, the network model is used to learn the deep features of the video scene and output the pixel-level classification results to achieve moving target detection. This method not only can adapt to complex video scenes of different sizes but also has a simple background extraction method, which effectively improves the detection speed.

Moving Target Detection Classifier for Airborne Radar Using SqueezeNet

Conventional moving target detection methods for airborne radar always need many training range units. To solve this problem, this paper transforms the target detection problem into a multi-classification problem. Firstly, the training dataset is constructed based on a small amount of training range units. Then, a multi-class classifier based on SqueezeNet is constructed. Finally, the trained classifier is used to extract the characteristics of the received space-time data for target detection and parameter estimation. Simulation results show that the SqueezeNet-based airborne radar moving target detection method proposed in this paper can effectively detect the target and estimate its distance, doppler frequency, and other parameters. Compared with the conventional space time adaptive processing method, the proposed method can significantly reduce the number of needed training range units. Compared with the existing target detection method based on classification, the proposed method can effectively improve the accuracy of target detection and parameter estimation.

Moving Target Detection Method of Three-Dimensional Image of Whip Leg Technique in Sanda

In the training of Chinese Wushu, Sanda has a history of more than 20 years. In the research of Sanda athletes, experts, scholars and so on, it integrates some foreign Sanda economic technology and constructs a set of relatively perfect technical system and theoretical framework, which makes Sanda become one of the international competitive events with rapid development. From the point of view of Sanda related theoretical research, it is not matched with the development of practice, which is obviously backward. In order to realize the scientific and effective promotion of Sanda, we must strengthen the treatment of this problem and enrich the relevant theory as soon as possible. In Sanda, whip leg is one of the key leg techniques, which is aggressive and can play a good role in attack and also an important score item in competition. The score rate of this part of skills in the economy is relatively high, but the utilization rate is not. It can be seen that there are some deficiencies and problems in the teaching and training of whip leg technology. In this regard, based on the perspective of sports biomechanics, this paper analyzes the technical action of whip leg with the help of three-dimensional image detection method, so as to provide some theoretical guidance for the training of whip leg technical action in Sanda.

Intelligent Monitoring Method of Short‐Distance Swimming Physical Function Fatigue Limit Mobile Calculation

The detection and classification of moving targets have always been a key technology in intelligent video surveillance. Current detection and classification algorithms for moving targets still face many difficulties, mainly because of the complexity of the monitoring environment and the limitations of target characteristics. Therefore, this article conducts corresponding research on moving target detection and classification in intelligent video surveillance. According to the Gaussian Mixture Background Model and Frame Difference Method, this paper proposes a moving target detection method based on GMM (Gaussians Mixture Model) and Frame Difference Method. This method first proposes a new image combination algorithm that combines GMM and frame difference method, which solves the problems of noise and voids inside the target caused by the fusion of traditional GMM and frame difference method. The moving target detection method can effectively solve the problems of incomplete moving target detection, target internal gap, and noise, and it plays a vital role in the subsequent moving target classification process. Then, the method adds image inpainting technology to compensate the moving target in space and obtain a better target shape. The innovation of this paper is that in order to solve the multiobject classification problem, a binary tree decision support vector machine based on statistical learning is constructed as a classifier for moving object classification. Improve the learning efficiency of the classifier, solve the competitive classification problem of the traditional SVM, and increase the efficiency of the mobile computing intelligent monitoring method by more than 70%.

Intelligent Video Surveillance Technology in Intelligent Transportation

Along with the strength of the country’s overall strength, the people’s pockets have become more and more popular, and there have been significant improvements in all aspects of life, especially in terms of travel methods. This reflects the increase in residents’ income, but it also brings huge traffic pressure. In the long run, traffic congestion is not only detrimental to urban development, but frequent traffic accidents threaten residents’ travel safety. Effective monitoring methods are essential to solving these problems, so it is necessary to carry out research on intelligent video monitoring technology in intelligent transportation. The purpose of this article is to solve the current situation of excessive traffic pressure in the city. Through the study of intelligent video surveillance technology in intelligent traffic, the use of constrained least squares algorithm to remove motion blur and apply Kalan filtering to the sharpening process is used to eliminate noise ambiguity and make a brief introduction to various classic moving target detection methods to realize real-time monitoring of intelligent traffic conditions and continuously adjust and verify the monitoring situation, and then establish intelligent video in intelligent traffic monitoring technology research system. The research results show that this kind of intelligent video surveillance technology research in intelligent transportation can effectively increase the awareness of intelligent video surveillance technology and improve the level of intelligent video surveillance technology. The data measurement time has been shortened by one hour, the aggregation time has been changed from three hours to two hours, and the analysis time has been shortened by half. Eased urban traffic road pressure and greatly reduced the incidence of traffic accidents, which is conducive to socialist harmony social construction.

A Hybrid Integration Method for Moving Target Detection With GNSS-Based Passive Radar

Global navigation satellite system (GNSS) based passive radar has been applied in the detection of moving targets. However, the low signal power of GNSS on the earth's surface limits the application of this technology for the long-range or low-observable target detection. Increasing the observation time can effectively improve the detection capability. But the target motion involves the range cell migration (RCM) and the Doppler frequency migration (DFM) over the long observation time, which results in the integration gain loss and lower the detection performance. This article proposes a new hybrid coherent and noncoherent integration method named the keystone transform and Lv's distribution. The proposed method not only compensate the RCM and the DFM but also provide coherent and noncoherent integration gains to increase the signal-to-noise ratio. The simulated results and the field trial results demonstrate that the detection performance of the proposed method is superior to the other two known moving target detection methods. And the analysis of the computational complexity shows that the proposed method and the other two methods are in the same order of O(N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> logN).

A Novel Moving Target Detection Method Based on RPCA for SAR Systems

Clutter background suppression and velocity estimation for moving targets are two critical problems in synthetic aperture radar-ground moving target indication (SAR-GMTI). A robust principal component analysis (RPCA) method is used to separate the sparse matrix of moving targets from the low-rank matrix of static backgrounds by using amplitude information in the image domain. However, the nonsparsity of the moving target echoes limits the performance of the RPCA in SAR-GMTI, and the velocity of the moving target cannot be estimated since the phase information is destroyed by the soft-thresholding operator in the RPCA process. To solve these problems, a novel moving target detection method based on RPCA (NRPCA) for SAR systems is proposed in this article. An atomic norm-based optimization program is first constructed to transform the data sparsity requirement into a moving target sparsity requirement. Although this optimization program is NP-hard, it is transformed to semidefinite programming by relaxation. Furthermore, accurate velocity estimation is performed using dual function theory and the alternating direction method of multipliers (ADMM) algorithm while the selection of the sparsity order k is avoided. Simulations and analyses based on experimental data illustrate the effectiveness of the proposed method.

Phase compensation transform for human detection with LFMCW radar

Efficiently accumulating the power of human echo using a linear frequency modulated continuous wave radar is difficult owing to the spread of micro-Doppler due to the varying velocity of human body motion. Consequently, the detection performance for human targets degrades. A coherent integration algorithm called phase compensation transform (PCT) is proposed to solve this issue. In PCT, the beat signal of a human target is derived for an LFMCW radar based on the Boulic model. The nonlinear phase characteristic of the beat signal caused by micro-Doppler from human motion is analyzed. The phase compensation signal is constructed to compensate the nonlinear phase from micro-Doppler during a coherent processing interval. The compensated range cell is fast Fourier transformed, and the Doppler spread resulting from periodic micro-Doppler is integrated. The performance of PCT is evaluated and compared with that of other methods. Simulation results show that PCT exhibits higher integration efficiency than moving target detection (MTD) and optimized nonlinear-phase methods. Experiment results from the LFMCW radar system demonstrate that the detection signal-to-noise ratio of PCT increases 5dB compared with that of MTD method. Besides, PCT can estimate human motion parameters simultaneously, including average radial velocity, gait frequency, and torso initial phase.

Fuzzy-based MTD

PurposeSynthetic aperture radar exploits the receiving signals in the antenna for detecting the moving targets and estimates the motion parameters of the moving objects. The limitation of the existing methods is regarding the poor power density such that those received signals are essentially to be transformed to the background ratio. To overcome this issue, fractional Fourier transform (FrFT) is employed in the moving target detection (MTD) process. The paper aims to discuss this issue.Design/methodology/approachThe proposed MTD method uses the fuzzy decisive approach for detecting the moving target in the search space. The received signal and the FrFT of the received signal are subjected to the calculation of correlation using the ambiguity function. Based on the correlation, the location of the target is identified in the search space and is fed to the fuzzy decisive module, which detects the target location using the fuzzy linguistic rules.FindingsThe simulation is performed, and the analysis is carried out based on the metrics, like detection time, missed target rate, and MSE. From the analysis, it can be shown that the proposed Fuzzy-based MTD process detected the object in 5.0237 secs with a minimum missed target rate of 0.1210 and MSE of 23377.48.Originality/valueThe proposed Fuzzy-MTD is the application of the fuzzy rules for locating the moving target in search space based on the peak energy of the original received signal and FrFT of the original received signal.

Adaptive Recognition of Motion Decomposition Image Based on Adaptive Principal Component Extraction Algorithm

Moving target detection and recognition methods are the foundation and key of modern intelligent video recognition systems. It combines advanced technologies in many fields such as image processing, pattern recognition, and artificial intelligence, and is a research hotspot in computer vision technology. Therefore, it is of great significance to study moving target detection and recognition algorithms. The non-local image extraction algorithm proposed in this paper uses an adaptive clustering method to perform fine cluster analysis on non-local image blocks with different feature types. Through the step-by-step principal component approximation method, we carefully find the features in each class. This progressive principal component approximation implements singular value hard threshold processing based on the Marchenko-Pastur (MP) theorem to select the main part of the feature, and uses special soft thresholds in the principal component transform domain to further improve the extraction performance. The Lower Bound-Based Within-Class Maximum Division (LBWCMD) is proposed, and this method is used as a preprocessing step of robust principal component analysis in moving target detection. This article applies LBWCMD to the video frame set based on the position information of the moving target, and the obtained frame subset meets the signal requirements of Robust Principle Component Analysis (RPCA) to the greatest extent. On this basis, we add frames with smaller motion amplitudes to each frame subset to increase the proportion of background pixels in each subset. Frame set division and low-rank decomposition realize the detection of moving targets under a unified framework. The detection rate of the proposed method is higher than that of the current popular methods in sports video data sets, and the detection accuracy is improved compared with the original RPCA method.

A Space-Time Reverberation Model for Moving Target Detection

In recent years, moving target detection methods based on low-rank and sparse matrix decomposition have been developed, and they have achieved good results. However, there is not enough interpretation to support the assumption that there is a high correlation among the reverberations after each transmitting pulse. In order to explain the correlation of reverberations, a new reverberation model is proposed from the perspective of scattering cells in this paper. The scattering cells are the subarea divided from the detection area. The energy fluctuation of a scattering cell with time and the influence of the neighboring cells are considered. Key parameters of the model were analyzed by numerical analysis, and the applicability of the model was verified by experimental analysis. The results showed that the model can be used for several simulations to evaluate the performance of moving target detection methods.

Moving target detection on an iron‐bridge using phase analysis for automotive FMCW radar

When driving on an iron-bridge, stationary iron-clutters cause disturbances in the detection of a moving target in an automotive radar system. In this Letter, the authors propose a moving target detection method that utilises phase similarity of the spatial domain on an iron-bridge. To analyse the spatial domain via the array antenna signal of the up- and down-chirp beat-frequencies, the frequency-modulated continuous wave (FMCW)-phased array radar signal was used. It is possible to match the up- and down-chirp beat-frequencies of each object, moving target, and stationary iron-clutters, by using the proposed method. Not only the moving target but also the ego-vehicle and moving target velocity are detected through this matching. The suitability of the proposed method is verified by on-road measurement data using a 77 GHz FMCW-phased array radar.

Research on Human Movement Target Recognition Algorithm in Complex Traffic Environment

With the increase in the total population of the society and the continuous increase in the number of trips, the traffic pressures faced by people are increasing. With the development and advancement of computer technology, the emergence of intelligent transportation provides a better way to solve the problem of effectively alleviating traffic pressure and reducing the incidence of traffic accidents. In recent years, intelligent traffic monitoring system, as one of the important branches in the field of intelligent transportation, has also received more and more attention. Among them, video-based moving target recognition technology involves theoretical knowledge in various fields such as artificial intelligence, image processing, pattern recognition and computer vision. It is an important means to realize “safe city” and “smart city” and a key technology for intelligent monitoring. Therefore, the research on human motion target recognition algorithm in complex traffic environment has important theoretical and practical value. In the field of intelligent traffic monitoring, the moving target detection and recognition effect of video images will have certain influence on the classification and behavior understanding of subsequent moving targets. In this paper, the commonly used moving target detection methods are studied first, and the convergence problem of the traditional Adaboost algorithm is improved. An Adaboost algorithm based on adaptive weight update is proposed, and then the support vector machine (SVM) is used. The algorithm identifies the detected moving target. Finally, through simulation experiments on the acquired video images, the results show that the proposed human motion target recognition algorithm based on adaptive weight update Adaboost and SVM has good feasibility and rationality.

Coherent integration method for random pulse repetition interval radar based on non‐uniform keystone transform and non‐uniform fast Fourier transform

In order to achieve weak target detection, the range migration and Doppler frequency spread induced by the target motion within the coherent processing interval should be compensated to focus the target energy in a certain range-Doppler cell. This article proposes a new integration method based on non-uniform Keystone transform and non-uniform fast Fourier transform (NUFFT) for random pulse repetition interval (PRI) radar. In this method, non-uniform Keystone transform is applied to correct the linear range migration. After that, the acceleration is estimated by dechirp process and the Doppler frequency spread is compensated. Finally, the target energy is accumulated coherently via NUFFT. Compared with the random PRI generalised Radon-Fourier transform (RPRI-GRFT) and the moving target detection method for random PRI radar, the proposed algorithm achieves a good balance between the detection performance and the computational cost. Simulation results are provided to demonstrate the effectiveness of the proposed method.

Three‐step ground moving target detection method for bistatic forward‐looking SAR

Detecting ground moving target in bistatic forward-looking synthetic aperture radar (BFSAR) has become increasingly important in military and civil field. Compared with pulse Doppler radar, the long observation time and high resolution of BFSAR will lead to three problems in the received signal: Doppler ambiguity, large range cell migration and Doppler cell migration, which tremendously aggravate the difficulty of ground moving target detection (GMTD). In this study, the authors propose a three-step ground moving target detection method for BFSAR to solve these problems and achieve good performance on GMTD. First, the bulk-deramp filtering and Keystone transform are exploited to suppress Doppler ambiguity and correct range cell migration, respectively. Then, time-division space–time adaptive processing is utilised for clutter suppression, which eliminates the effect of the Doppler cell migration. Finally, modified Wigner–Ville distribution method is applied to further improve the signal to clutter and noise ratio and the moving target can be detected in the integration domain. Simulations are carried out to demonstrate the effectiveness of the proposed method.

Motion classification for radar moving target via STFT and convolution neural network

Radar moving target detection (MTD) technology is a key technology in the field of radar signal processing. The MTD method of radar has a better performance when applied to uniform motion detection although it has limited performance in other aspects, and it is also difficult to distinguish the type of the moving target. This article presents a new method for detecting and classifying moving targets based on convolutional neural network, which uses the convolutional neural network to learn the motion characteristics of the moving target echo so that realises the detection and classification of moving targets. At first, the authors model the moving target echo. Then the short-time Fourier transform (STFT) is used to perform time–frequency analysis. The obtained time–frequency graph is used as the input of the convolutional neural network which can automatically learns moving target features through training. Finally, the authors use the trained convolutional neural network model to detect and classify moving targets. The simulation verifies that the method has a great improvement in the detection accuracy rate compared with the traditional MTD.

Moving target coherent integration method for non‐coherent short pulse radar

The microwave power source of non-coherent short pulse (NCSP) radar is a relativistic backward wave oscillator, which is a kind of vacuum electronics device. The moving target coherent integration method for the NCSP radar is a challenging task. Firstly, due to the mechanism of generating microwave pulses, the centre frequency and initial phase of the transmitting signal in the NCSP radar are random, and the envelope of the microwave pulse also has jitter. Therefore, the well-known moving target detection method cannot be carried out straightforwardly by the NCSP radar. Secondly, range cell migration (RCM) effects induced by target motion would result in serious integration performance loss. In this study, a method based on Radon–Fourier transform is proposed to the scenario of moving target coherent integration for the NCSP radar. Not only the non-coherent factors of transmitting signal but also RCM effects have been compensated carefully. Also, the staggered pulse repetition frequency technique is applied to overcome blind speed sidelobes effect. Numerical experiments are performed to demonstrate the effectiveness of the proposed method.

A New Maritime Moving Target Detection and Tracking Method for Airborne Forward-looking Scanning Radar.

Maritime moving target detection and tracking through particle filter based track-before-detect (PF-TBD) has significant practical value for airborne forward-looking scanning radar. However, villainous weather and surging of ocean waves make it extremely difficult to accurately obtain a statistical model for sea clutter. As the likelihood ratio calculation in PF-TBD is dependent on the distribution of the clutter, the performance of traditional distribution-based PF-TBD seriously declines. To resolve these difficulties, this paper proposes a new target detection and tracking method, named spectral-residual-binary-entropy-based PF-TBD (SRBE-PF-TBD), which is independent from the prior knowledge of sea clutter. In the proposed method, the likelihood ratio calculation is implemented by first extracting the spectral residual of the input image to obtain the saliency map, and then constructing likelihood ratio through a binarization processing and information entropy calculation. Simulation results show that the proposed method had superior performance of maritime moving target detection and tracking.