Problem Of Target Recognition Research Articles

An open-set recognition method for ship radiated noise signal based on graph convolutional neural network prototype learning

The underwater acoustic perception system often undertakes multi-class ship recognition tasks. As the underwater acoustic recognition environment becomes increasingly complex, underwater acoustic perception systems often face various interference, such as unknown ships radiated noise signals and decoy signals. This poses higher requirements on the robustness and recognition capabilities of underwater acoustic target recognition methods. In this paper, we transform the problem of underwater acoustic target recognition in this scenario into an open-set recognition problem. We design a deep learning model based on graph convolutional neural networks, propose a graph embedding method for time-domain ships radiated noise signals, and propose a fully parameterized prototype learning framework. We simulate decoy signals in real sea areas, and all the data used in the experiments are actual collected data. The fully parameterized prototype learning framework based on a data-driven approach can not only effectively resist interference from unknown ship-radiated noise signals and decoy signals, but also accurately identify multi-class target ship-radiated noise signals. Ultimately, our method achieves end-to-end open-set recognition of ship-radiated noise signals.

Non-cooperative target recognition and relative motion estimation with inertial measurement unit assistance

This study investigated the problems of non-cooperative target recognition and relative motion estimation during spacecraft rendezvous maneuvers. A structure integrating an Inertial Measurement Unit (IMU) and a visual camera was presented. The angular velocity output of the IMU was used to calculate the motion trajectories of star points in multiple image frames, which can highlight the motion of non-cooperative targets with respect to the image background to improve the probability of target recognition. To solve the problem of target misidentification caused by new star points entering the field of view, a target-tracking link based on IMU prediction was introduced to track the position of the target in the image. Furthermore, a measurement model was constructed using the line-of-sight vector generated from target recognition, and the relative motion state was estimated using a Huber-based non-linear filter. Semi-physical and numerical simulations were performed to evaluate the effectiveness and efficiency of the proposed method.

GLRT‐based detection of targets composed of distributed scattering centres

AbstractAttributed scattering problems have been found to be helpful in inverse synthetic aperture radar (ISAR) imaging and target recognition problems. In this model, the scattering centres are divided into two categories: localised and distributed. Localised scattering centres are those that are concentrated in a small area, while distributed scattering centres are spread out over a larger area. Several methods have been proposed to estimate the scattering centres which aim to accurately identify the location and characteristics of the scattering centres. However, detecting a distributed scattering centre remains a challenging task. A novel technique is proposed based on sparse signals to improve the detection of distributed scattering centres from localised ones. This technique takes advantage of the sparsity of the signals to accurately identify the location of the distributed scattering centres. Experimental results demonstrate the superiority of algorithm in detecting distributed scattering centres. This improved detection capability has significant implications for ISAR imaging and target recognition problems.

An Acoustic Array Sensor Signal Recognition Algorithm for Low-Altitude Targets Using Multiple Five-Element Acoustic Positioning Systems with VMD

To solve the problem in target acoustic signal processing and recognition when the target flies at a low altitude based on the acoustic positioning system, which is often affected by external interference and brings false information, this paper proposes a target signal processing and recognition algorithm for low-altitude target acoustic positioning based on variational modal decomposition and the test method of multiple five-element acoustic arrays. This algorithm uses VMD to decompose the target signal into modal components with different central frequencies and then performs wavelet threshold processing on the low-frequency part of the signal. After determining the remaining signal components and the low-frequency part’s threshold, the residual component is reconstructed. Based on the test principle and calculation model of the five-element acoustic positioning system, following processing of the low-altitude target acoustic positioning signal using variational modal decomposition, the cross-correlation function method is introduced to perform correlation operations on the basic array of five acoustic sensors and then obtain the time value and time difference of the target acoustic information in each acoustic sensor, ultimately determining the spatial position of the target. Finally, we used the data fusion processing method for target coordinates in multi-acoustic basic arrays to determine the actual target position. By comparing the results obtained using the high-speed camera method with those of the proposed approach, it was found that the average error in the test area of 100 × 100 m was less than 1 m.

Simulation and parameter extraction of flying‐bird motion echo based on laser micro‐Doppler effect

AbstractTo solve the problem of target recognition for flying‐birds and biomimetic flapping‐wing unmanned aerial vehicles (UAVs), a rectangular‐trapezoidal composite scattering model for lidar detection of bird wings has been proposed and established. According to the model, the influence of different line‐of‐sight angles on the time–frequency distribution characteristics of laser micro‐Doppler under the same flapping wing motion was simulated and studied. Then, the flapping‐wing frequency and maximum frequency shift are extracted based on the two‐dimensional Fourier transform and the least squares method. When signal noise ratio = 5 dB, the accuracy rate of flapping‐wing frequency extraction can reach at 100%, the maximum relative error of frequency shift for inner contour extraction is less than 3.0%, and the maximum relative error of frequency shift for outer contour extraction is less than 7.2%.

Quadruplet depth-wise separable fusion convolution neural network for ballistic target recognition with limited samples

Radar high-resolution range profile (HRRP) plays a crucial role in ballistic target recognition due to its simplicity and fast computation. Convolutional neural networks (CNNs), a popular deep learning approach, are widely used in radar automatic target recognition (RATR) based on HRRP, thanks to their excellent feature extraction capabilities. However, current research mainly focuses on integrating branch networks like recurrent neural networks and attention mechanisms into vanilla CNNs to enhance feature extraction. Limited research exists on reducing the computational complexity of vanilla CNNs when applied to HRRP-based RATR, as well as target recognition problems with limited samples. To address these issues, this paper proposes a lightweight depth-wise separable fusion CNN (DSFCNN) for ballistic target HRRP recognition. The DSFCNN reduces the computational complexity of vanilla CNNs while improving recognition accuracy. Furthermore, we introduce the sample-level fitting and class-level distinguishability (SFCD) loss for limited-sample ballistic target HRRP recognition. The SFCD loss ensures better fitting performance of all samples, as well as higher intra-class compactness and inter-class separability, enhancing the distinguishability of embedded features in limited-sample conditions. During the training phase, the SFCD loss is equally applied to four DSFCNNs with shared structures and parameters, resulting in a model known as Quadruplet DSFCNN (QDSFCNN). However, during the testing phase, only one DSFCNN is utilized. Additionally, we propose the parameter quantity shifting-fitting performance (PQS-FP) coordinate system to elucidate the performance disparities between DSFCNN and vanilla CNNs in our experiments. PQS-FP offers inspirational insights into the relations among parameter quantity, fitting performance, and recognition performance. Extensive experimental results demonstrate the robustness and effectiveness of QDSFCNN for ballistic target recognition in various limited-sample scenarios.

An iterative mass-to-drag ratio estimation method for ballistic targets

Ballistic target tracking and identification is a key task for missile defense systems. A vital problem in target track and recognition is how to estimate the target mass-to-drag ratio fast and accurately. Based on the constraints of energy conservation, this paper proposes an iterative mass-to-drag ratio estimation method. This algorithm first estimates the position and velocity of the target using unscented Kalman filter, and then estimates the mass-to-drag ratio of the target by using energy conservation. By iterating, the final value is outputted. The simulation results show that this method can estimate the mass-to-drag ratio of the target at a high altitude, which provides a precise way for mass-to-drag ratio estimation.

Application of Target Recognition and Tracking in Agricultural Picking Robots

Target tracking in agricultural picking robots has a huge impact in the field of computer vision, and high-quality picking recognition and tracking algorithms are the basis for solving many problems such as inaccurate localization in agricultural picking. The problem of target recognition and tracking research is to identify and track the trajectories of different targets in video sequences. This paper mainly summarizes and analyzes the latest proposed algorithmic models about deep learning on target recognition and tracking and its application in agricultural picking robots. By analyzing various problems of current target recognition, the improvement of algorithms for different problems is summarized, and finally some possible future research directions are also proposed.

Transductive Prototypical Attention Reasoning Network for Few-Shot SAR Target Recognition

Deep learning-based synthetic aperture radar (SAR) automatic target recognition (ATR) algorithms have achieved outstanding performance under the condition of hundreds or thousands of training samples in recent years. Nevertheless, it is often rare to acquire great quantities of target samples in real SAR application scenarios. This article proposes a novel ATR method called transductive prototypical attention reasoning network (TPARN) to solve the problem of SAR target recognition with only a few training samples. To be specific, a region awareness-based feature extraction model is first developed, which can effectively focus on the target region of interest and suppress the background clutter by embedding direction-aware and position-sensitive information to extract more transferable knowledge. To heighten the discrimination of the sample features, a cross-feature spatial attention module is then proposed following the feature embedding model. Finally, a transductive prototype reasoning method is presented to realize the identity reasoning of the target, which can continuously update each class prototype with training samples and test samples together, thereby improving the classification accuracy. In addition, a marginal adaptive hybrid loss is proposed to obtain a discriminative feature embedding space with intra-class compactness and inter-class divergence, aiming to facilitate subsequent target identity reasoning. Extensive experiments on the moving and stationary target acquisition and recognition (MSTAR) benchmark dataset reveal that the proposed method outperforms some state-of-the-arts under different few-shot SAR ATR tasks.

A deep learning method for X-ray image safety detection: YOLO-T

There are many problems in X-ray image dangerous goods target recognition with existing technology, such as low degree of automation, slow detection time, easy to misjudge under occlusion interference, etc. Based on the above problems, this paper proposes a multi-objective intelligent security inspection method for X-ray images based on the YOLO-T deep learning network. By adding the optimized Transformer structure to the YOLO architecture, this method can better solve the above problems. In order to better carry out the experiment, we proposed a set of X-ray safety detection data set GDXray-Expanded containing multiple categories of dangerous goods, and tested several versions of the deep learning network model of the YOLO series on this basis. Experiments show that the existing YOLO series algorithms still cannot solve the problem that dangerous goods in X-ray images are easy to be misjudged under occlusion interference. The YOLO-T method proposed in this paper solves this problem well, and in the big data set test, the maximum mAP can reach 97.73%, which is 7.66%, 16.47%, and 7.11% higher than the three methods of YOLO v2, YOLO v3, and YOLO v4 respectively, and has achieved the most competitive performance in the detection of seven categories of dangerous goods. To sum up, the YOLO-T network proposed in this paper mainly solves a series of problems in the field of dangerous goods target recognition and detection in X-ray security inspection images and has a high engineering application prospect in the field of X-ray security inspection.

A Deep Learning Model Applied to Optical Image Target Detection and Recognition for the Identification of Underwater Biostructures

Objective: We propose a deep-learning-based underwater target detection system that can effectively solve the problem of underwater optical image target detection and recognition. Methods: In this paper, based on the depth of the underwater optical image target detection and recognition and using a learning model, we put forward corresponding solutions using the concept of style migration solutions, such as training samples. A lack of variability and poor generalization of practical applications presents a challenge for underwater object identification. The UW_YOLOv3 lightweight model was proposed to solve the problems of calculating energy consumption and storage resource limitations in underwater application scenarios. The detection and recognition module, based on deep learning, can deal with the degradation process of underwater imaging by embedding an image enhancement module into the detection and recognition module for the joint tuning and transferring of knowledge. Results: The detection accuracy of the UW_YOLOv3 model designed in this paper outperformed the lightweight algorithm YOLOV3-TINY by 7.9% at the same image scale input. Compared with other large algorithms, the detection accuracy was lower, but the detection speed was much higher. Compared with the SSD algorithm, the detection accuracy was only 4.7 lower; the speed was 40.9 FPS higher; and the rate was nearly 16 times higher than Faster R-CNN. When the input scale was 224, although part of the accuracy was lost, the detection speed doubled, reaching 156.9 FPS. Conclusion: Based on our framework, the problem of underwater optical image target detection and recognition can be effectively solved. Relevant studies have not only enriched the theory of target detection and glory, but have also provided optical glasses with a clear vision for appropriate underwater application systems.

CNN-Based Multiterrain Moving Target Recognition Model for Unattended Ground Sensor Systems

In recent years, many deep learning algorithms based on seismic signals have been proposed to solve the moving target recognition problem in unattended ground sensor systems. Despite the excellent performance of these deep networks, most of them can only be deployed on cloud-based devices and cannot be deployed on low-power hardware devices due to the large network size. Second, since seismic signals are affected by the terrain, employing only seismic signals as reconnaissance means for unattended ground sensors cannot achieve multiterrain-type adaptability. In response, this paper proposes an MFC-TinyNet method facing a multiterrain. The method adds depthwise separable convolutional layers to the network, which effectively reduces the size of the network while keeping the target recognition accuracy constant, and solves the problem that the model is difficult to deploy on low-power hardware. It also uses the Mel-frequency spectrum feature extraction method to fuse sound and seismic signals to improve the accuracy of the model’s moving target recognition on a multiterrain. Experiments demonstrate that the method can combine the two advantages of the small network model and multiterrain applicability.

Multiview Fusion 3D Target Information Perception Model in Nighttime Unmanned Intelligent Vehicles

Unmanned technology is an important development project of today’s cutting-edge science and technology, which has a significant impact on social and economic development, national defense construction, and scientific and technological development. The rapid development of industrial information technology has driven the unmanned intelligent vehicle system to innovate and gradually enter the public’s view, and at the same time, the driving safety of unmanned intelligent vehicles is also widely concerned. Target information perception system is the foundation of unmanned system and the fundamental guarantee of safety and intelligence of unmanned vehicles. There are three key problems of target recognition in unmanned driving, namely, target classification, localization, and attitude determination. In the implementation of a networked virtual environment system, a flexible and complete perception model is needed as the guiding model of the system. Since 3D point cloud data can provide more spatial information than 2D RGB image data, it is more beneficial to determine the target category, position, and pose in 3D. In this paper, based on the existing research of unmanned intelligent vehicle perception system, we combine the application of fusion of 3D target information perception model and develop a nighttime unmanned system based on multiview fusion of 3D target information perception model. This system can simultaneously perform the detection of multiple categories of objects and predict the center point, length, width, height, and orientation of the objects, so that the unmanned car can sense the location of the surrounding objects when driving in the actual scene.

A Multi-Rotor Drone Micro-Motion Parameter Estimation Method Based on CVMD and SVD

It is of great significance to detect drones in airspace due to the substantial increase and regrettable misuse in the consumer market. In this paper, we establish a micro-motion theoretical model of a drone and analyze the micro-Doppler signature of rotor targets and the flicker mechanisms of the multi-rotor targets. Hence, for the target recognition problem of multi-rotor drones, a multi-rotor target micro-Doppler parameter estimation method is proposed. Firstly, a signal frequency domain segmentation method is proposed based on the complex variational mode decomposition (CVMD) to separate the high-frequency part of the high-frequency flicker in the frequency domain. Secondly, for the signal after frequency domain segmentation, a flicker time domain position method based on singular value decomposition (SVD) is proposed. Finally, by integrating CVMD frequency domain segmentation and SVD time domain positioning, the reconstruction of multi-rotor target scintillation at different speeds is realized, and the micro-motion parameters of rotor blades are successfully estimated. The simulation results show that the method has high accuracy in estimating the micro-motion parameters of a multi-rotor, which makes up for the shortage of the traditional method in estimating the micro-motion parameters of the multi-rotor target.

A multi-view SAR target recognition method using feature fusion and joint classification

ABSTRACT To handle synthetic aperture radar (SAR) image target recognition problem, a multi-view method is proposed. For the multi-view SAR images be recognized, they are first clustered based on the correlation coefficients and divided into several view sets. Afterwards, for the view set containing two or more images, the multiset canonical correlation analysis (MCCA) is employed to fuse them as a single feature vector. For the view set with a single image, its corresponding feature vector is directly used. Finally, the joint sparse representation is used to characterize and classify the feature vectors from different view sets and determine the target label of the multi-view SAR images. Experiments and analysis on the moving and stationary target acquisition and recognition (MSTAR) dataset show that the proposed method can achieve an average recognition rate of 99.42% for 10 types of targets under the standard operating condition (SOC). Its performance is also better than several reference methods under the extended operating conditions (EOC) including noise interference and target occlusion.

Parallel CNN Network Learning-Based Video Object Recognition for UAV Ground Detection

Video object recognition for UAV ground detection is widely used in target search, daily patrol, environmental reconnaissance, and other fields. So, we propose the novel parallel deep learning network with the ability of the global and local joint feature extraction for the UAV video target detection. This paper focuses on solving the problems of feature extraction and target background discrimination required by target discovery to realize target discovery. Break through the key problems of real-time target recognition, such as multiscale targets, high background complexity, many small targets, dense target arrangement, and multidirection, and put forward an optimized network scheme, aiming at the problem of multiscale of image target and aiming at the problem of large change of target scale in image. In the network, the corresponding targets with different sizes and different aspect ratios are matched to make the different targets match the closest, and then, the position of the detection box is fine-tuned by regression. For the special problem of image viewing angle and for the rotation invariance of the airborne down looking image of the target, the usual solution is through data enhancement; that is, through the rotation transformation of the training data, the neural network can learn the rotation invariance of the target. Aiming at the problem of multi-directional image target and aiming at the problems of large target aspect ratio, large target tilt angle, and changeable direction in the target, we propose to use the tilt detection frame instead of the ordinary rectangular detection frame. Aiming at the problem of dense arrangement of image targets and aiming at a large number of densely arranged targets in the image, a feature refining module is proposed, which can effectively improve the detection performance of the detector for densely arranged targets. The experimental results shows that the proposed algorithm achieves more than 10% on the target detection accuracy with focal length change of 1-10 times. The detection accuracy meets the requirements of practical application.

Target recognition of heterogeneous fusion images based on improved YOLO model deep learning algorithm

Aiming at the problem of target recognition of heterogeneous fusion images, the algorithm flow and network structure of the YOLO convolutional neural network model are analyzed and explained based on the deep learning of the convolutional neural network. Combining the characteristics of the fusion image, the interface, structure and parameters of the YOLOv3 convolutional neural network are improved and adjusted while the recognition effect is guaranteed, the YOLO model is streamlined, and the training is successfully implemented and the training weights are output. The algorithm is effective and stable, and the recognition speed is faster.

Fine-grained Recognition of Ships Under Complex Sea Conditions

Abstract For the traditional deep learning cannot solve the fog, coastal background interference, and the difficulty of small ships recognition, a multi-scale deep learning training model is proposed in this paper. Based on Faster R-CNN, this paper uses guided filtering to remove fog, as well as combined with negative sample enhancement learning to train the model, thus solving recognition of ship in complex sea conditions. And with multi-scale training strategy, the multi-scale ship samples are produced and sent to the network for training, so as to solve the problem of small target recognition. The experimental results show that compared with the Faster R-CNN, the precision and recall of our method increase by 6.43% and by 4.68% respectively. It solves the difficulty of ships recognition under complex sea conditions and small ship recognition that cannot be solved by traditional deep learning methods, the trained model has good generalization ability and robustness.

End-to-End Recognition of Similar Space Cone–Cylinder Targets Based on Complex-Valued Coordinate Attention Networks

Except for the slight difference of micromotion parameters, some decoys and warheads have the same geometry and micromotion form. As a result, recognition of similar space cone–cylinder targets is one of the difficult problems in ballistic target recognition. In recent years, due to the good effect of deep neural networks (DNNs) in optical target recognition, space cone–cylinder target recognition methods based on DNN have attracted wide attention. However, these DNN-based methods only recognize the space cone–cylinder targets with different shapes and different micromotion forms. Moreover, these methods require some time-consuming preprocessing operations, which need to observe the target for at least one micromotion period. To recognize similar space cone–cylinder targets, we propose a complex-valued coordinate attention networks (CV-CANets)-based end-to-end recognition method. Firstly, we establish the signal model of space cone–cylinder targets. Secondly, we propose CV-CA blocks by transforming the coordinate attention mechanism into the complex-valued domain. Then, we construct CV-CANet based on the proposed CV-CA blocks. Finally, the proposed CV-CANet is trained and tested by the narrowband radar echo data, which is generated by electromagnetic calculation. Compared with the convolutional neural network (CNN)-based recognition methods, the proposed method can not only recognize the similar space cone–cylinder targets but also is superior in terms of time cost and observation requirement. Extensive experiments validate that the proposed recognition method is effective when the targets only have a slight difference on the precession angular frequency and the observation time is less than half a period.

Combination of ELM and BEMD for Target Recognition of SAR Images

For the problem of target recognition of synthetic aperture radar (SAR) images, a method based on the combination of bidimensional empirical mode decomposition (BEMD) and extreme learning machine (ELM) is proposed. BEMD performs feature extraction for SAR images, producing multi-layer bidimensional intrinsic mode functions (BIMF). These BIMFs covey the discrimination of the original target while effectively eliminating the noises. ELM conducts the classification of each BIMF with high efficiency and robustness. Finally, the decisions from different BIMFs are fused using a linear weighting strategy to reach a reliable decision on the target label. The proposed method compensates the relatively low adaptivity of ELM to noise corruption by BEMD feature extraction. Moreover, the multi-layer BIMFs provide more discriminative information for correct decision. Hence, the overall recognition performance can be improved. As an efficient recognition algorithm, the proposed method can be used in an embedded system for wide applications. Experiments are designed and implemented on the moving and stationary target acquisition and recognition (MSTAR) dataset. The proposed method is tested under both the standard operating condition (SOC) and extended operating conditions (EOCs). The results reflect its effectiveness and robustness via quantitative comparisons.