Target Detection Performance Research Articles

TR-YOLO: A pig detection network based on YOLO V5n by combining self attention mechanism and large convolutional kernel

Under the highly valued environment of intelligent breeding, rapid and accurate detection of pigs in the breeding process can scientifically monitor the health of pigs and improve the welfare level of pigs. At present, the methods of live pig detection cannot complete the detection task in real time and accurately, so a pig detection model named TR-YOLO is proposed. Using cameras to collect data at the pig breeding site in Rongchang District, Chongqing City, LabelImg software is used to mark the position of pigs in the image, and data augmentation methods are used to expand the data samples, thus constructing a pig dataset. The lightweight YOLOv5n is selected as the baseline detection model. In order to complete the pig detection task more accurately, a C3DW module constructed by depth wise separable convolution with large convolution kernels is used to replace the C3 module in YOLOv5n, which enhances the receptive field of the whole detection model; a C3TR module constructed by Transformer structure is used to extract more refined global feature information. Contrast with the baseline model YOLOv5n, the new detection model does not increase additional computational load, and improves the accuracy of detection by 1.6 percentage points. Compared with other lightweight detection models, the new detection model has corresponding advantages in terms of parameter quantity, computational load, detection accuracy and so on. It can detect pigs in feeding more accurately while satisfying the real-time performance of target detection, providing an effective method for live monitoring and analysis of pigs at the production site.

Remote Sensing Micro-Object Detection under Global and Local Attention Mechanism

With the rapid advancement of technology, satellite and drone technologies have had significant impacts on various fields, creating both opportunities and challenges. In areas like the military, urban planning, and environmental monitoring, the application of remote sensing technology is paramount. However, due to the unique characteristics of remote sensing images, such as high resolution, large-scale scenes, and small, densely packed targets, remote sensing object detection faces numerous technical challenges. Traditional detection methods are inadequate for effectively detecting small targets, rendering the accurate and efficient detection of objects in complex remote sensing images a pressing issue. Current detection techniques fall short in accurately detecting small targets compared to medium and large ones, primarily due to limited feature information, insufficient contextual data, and poor localization capabilities for small targets. In response, we propose an innovative detection method. Unlike previous approaches that often focused solely on either local or contextual information, we introduce a novel Global and Local Attention Mechanism (GAL), providing an in-depth modeling method for input images. Our method integrates fine-grained local feature analysis with global contextual information processing. The local attention concentrates on details and spatial relationships within local windows, enabling the model to recognize intricate details in complex images. Meanwhile, the global attention addresses the entire image’s global information, capturing overarching patterns and structures, thus enhancing the model’s high-level semantic understanding. Ultimately, a specific mechanism fuses local details with global context, allowing the model to consider both aspects for a more precise and comprehensive interpretation of images. Furthermore, we have developed a multi-head prediction module that leverages semantic information at various scales to capture the multi-scale characteristics of remote sensing targets. Adding decoupled prediction heads aims to improve the accuracy and robustness of target detection. Additionally, we have innovatively designed the Ziou loss function, an advanced loss calculation, to enhance the model’s precision in small target localization, thereby boosting its overall performance in small target detection. Experimental results on the Visdrone2019 and DOTA datasets demonstrate that our method significantly surpasses traditional methods in detecting small targets in remote sensing imagery.

Multiscale Feature Extraction U-Net for Infrared Dim- and Small-Target Detection

The technology of infrared dim- and small-target detection is irreplaceable in many fields, such as those of missile early warning systems and forest fire prevention, among others. However, numerous components interfere with infrared imaging, presenting challenges for achieving successful detection of infrared dim and small targets with a low rate of false alarms. Hence, we propose a new infrared dim- and small-target detection network, Multiscale Feature Extraction U-Net for Infrared Dim- and Small-Target Detection (MFEU-Net), which can accurately detect targets in complex backgrounds. It uses the U-Net structure, and the encoders and decoders consist of ReSidual U-block and Inception, allowing rich multiscale feature information to be extracted. Thus, the effectiveness of algorithms in detecting very small-sized targets can be improved. In addition, through the multidimensional channel and spatial attention mechanism, the model can be adjusted to focus more on the target area in the image, improving its extraction of target information and detection performance in different scenarios. The experimental results show that our proposed algorithm outperforms other advanced algorithms in detection performance. On the MFIRST, SIRST, and IRSTD-1k datasets, we achieved detection rates of 0.864, 0.962, and 0.965; IoU values of 0.514, 0.671, and 0.630; and false alarm rates of 3.08 × 10−5, 2.61 × 10−6, and 1.81 × 10−5, respectively.

Real time object tracking using deep learning and OpenCV

This article summarizes the background and application fields of Open-Source Computer Vision Library (OpenCV) and deep learning and conducts research based on their object detection and tracking applications. The model algorithm selected in this article is the Convolutional Neural Networks (CNN) algorithm. CNN algorithm can be used for object detection in real-time scenarios. Moreover, the CNN algorithm has good credibility. The Python program developed based on CNN algorithm can effectively achieve real-time object tracking. The model shows good detection and tracking performance for trained targets and can be further applied in more specific scenarios in the future. Because deep learning can process large-scale data and recognize complex patterns, it can automatically learn and extract advanced features. Combining the two can achieve faster and more accurate detection and tracking of target objects. After having a large enough training sample size, it can detect and track the specified object more accurately. However, it is precisely due to the enormous sample size required for deep learning that there are still some difficulties in applying it to real-time object tracking. This article first discusses the use of supervised learning methods for deep understanding. When the input sample capacity is large enough, the machine can better reflect the real-time detection and tracking of the target object. But the time required will significantly increase. This issue still needs to be addressed in the subsequent research process.

Interference Suppression for an FM-Radio-Based Passive Radar via Deep Convolutional Autoencoder

Recently, the passive bistatic radar (PBR) which exploits frequency modulation (FM) radio transmitters as illuminators, has witnessed widespread usage owing to its various advantages. However, the characteristics of FM-radio-based PBR result in interference components in the range–Doppler (RD) map, which may increase false alarms. Therefore, this study proposes a method for suppressing interference components using a deep learning approach. The two main contributions of this study are as follows. First, a convolutional autoencoder model capable of effectively suppressing interference in the RD map of PBR was proposed. Second, a synthetic RD map dataset generation method that can enable the autoencoder to operate robustly in PBR in a real environment was presented. Further, a performance comparison between the proposed method and existing methods using simulated data proved that the deep learning-based method exhibited superior target detection performance. Furthermore, using the data recorded by the PBR in a real environment, the proposed autoencoder model was shown to effectively suppress interference components in a real interference environment.

Not even 6 dB: Gaussian quantum illumination in thermal background

In analyses of target detection with Gaussian state transmitters in a thermal background, the thermal occupation is taken to depend on the target reflectivity in a way which simplifies the analysis of the symmetric quantum hypothesis testing problem. However, this assumption precludes comparison of target detection performance between an arbitrary transmitter and a vacuum state transmitter, i.e. ‘detection without illumination’, which is relevant in a bright thermal background because a target can be detected by its optical shadow or some other perturbation of the background. Using a target-agnostic thermal environment leads to the result that the oft-claimed 6 dB possible reduction in the quantum Chernoff exponent for a two-mode squeezed vacuum transmitter over a coherent state transmitter in high-occupation thermal background is an unachievable limiting value, only occurring in a limit in which the target detection problem is ill-posed. Further analyzing quantum illumination in a target-agnostic thermal environment shows that a weak single-mode squeezed transmitter performs worse than ‘no illumination’, which is explained by the noise-increasing property of reflected low-intensity squeezed light.

Innovative approaches to suppress non-specific adsorption in molecularly imprinted polymers for sensing applications

Molecularly imprinted polymers (MIPs) are synthetic antibodies developed to bind selectively with specific molecules. They function through a particular recognition process involving their cavities and functional groups. Nevertheless, functional groups located outside these cavities are the main cause of non-specific molecule binding, thus reducing the effectiveness of MIPs in sensing applications. This work focused on enhancing the selectivity and performance of MIPs through electrostatic modification with surfactants. The study investigates the use of two surfactants, namely sodium dodecyl sulfate (SDS) and cetyl trimethyl ammonium bromide (CTAB), to eliminate non-specific adsorption in MIPs. The binding isotherms of the target molecule sulfamethoxazole (SMX) on MIPs and non-imprinted polymers (NIPs) were analyzed, showing higher adsorption capacity of MIPs due to the specific cavities. The modification with SDS or CTAB effectively eliminated non-specific adsorption in MIPs. The kinetic adsorption behavior further demonstrated the efficacy of MIP+--SDS/CTAB in the selective adsorption of SMX. Calibration curves showcase the methodology's analytical capabilities, achieving low limit of detection for SMX 6 ng mL−1 using MIP +-SDS. The stability study confirmed that the developed MIP +/--SDS/CTAB remains stable even at high temperatures, demonstrating its suitability for on-site applications. The methodology was successfully applied to detect SMX in milk and water samples, achieving promising recoveries. Overall, the electrostatic modification of MIPs with surfactants emerges as a valuable strategy for enhancing selectivity and performance in target molecule recognition and detection.

Robust SR-STAP algorithms in partly calibrated arrays for airborne radar

Sparse recovery based space-time processing (SR-STAP) techniques are capable of achieving satisfactory clutter suppression and target detection performance, even with limited training samples. The majority of existing approaches are developed under the assumption of ideal uniform linear arrays, where no imperfections are present. In this study, we consider the robust STAP problem in the context of airborne partly calibrated array (PCA) composed of several well calibrated subarrays with unknown inter-subarray gain-phase and displacement errors. Initially, the space-time signal for airborne STAP radar based on PCAs is modeled. Subsequently, the authors propose two robust SR-STAP algorithms, considering both grid-based and gridless methods, leveraging the block-sparsity and low rank structural characteristics inherent in the signal model. Extensive numerical experiments have shown the significant performance benefits including small sample applicability and resilience to inter-subarray gain-phase and displacement errors.

Comprehensive visual information acquisition for tomato picking robot based on multitask convolutional neural network

The tomato picking robot's vision system faces two difficult tasks: precise tomato pose acquisition and stem location. Tomato pose and stem location can help determine the end effector pose and achieve collision-free picking. To realise efficient crop picking, the tasks of target location, pose detection, and obstacle semantic segmentation should be completed in one model to obtain comprehensive visual information. Therefore, the multitask convolutional neural network YOLO-MCNN is proposed, a new method to complete the above tasks in one model. By fusing multi-scale features and determining the optimal locations for the semantic segmentation branch, four strategies are proposed for enhancing the segmentation ability. The experiment results show that fusing the semantic segmentation branch with the second layer of shallow feature maps and placing the branch after the 17th layer can result in the best segmentation performance. Fusing shallow feature maps improves small target detection while merging multi-scale feature maps enhances semantic segmentation performance. Moreover, ablation experiments are conducted to understand the influence between multitask convolutional and single task networks. It proves that running multiple tasks on the same backbone network does not affect their performance. The YOLO-MCNN's target detection performance F1 is 87.8%, the semantic segmentation performance mIoU is 74.8%, the keypoint detection performance dlmk is 6.95 pixels, the network size is 15.2 MB, and the inference speed is 19.9ms. Compared with other target detection and semantic segmentation networks, it shows that the comprehensive performance of the YOLO-MCNN is the best. The method provides theoretical foundation for constructing multitask convolutional neural networks.

Information theory based waveform design for RIS-aided multifunction RF systems

This paper studies the waveform design problem for reconfigurable intelligent surface (RIS) aided multifunction radio frequency (MFRF) systems, which enables detection, communication, and electronic attack simultaneously. The design purpose is to maximize the relative entropy while ensuring effective communication and jamming functionalities. To improve the practicability of the MFRF system, a constant-modulus constraint is imposed on the transmit waveforms. A cyclic method is proposed to iteratively optimize the transmit waveforms and the RIS coefficients. Specifically, to tackle the resultant non-convex problems, we resort to the minorization-maximization based method to transform the objective function into a quadratic form, then leverage alternating direction method of multipliers to seek efficient solutions. Numerical results verify the effectiveness of the proposed algorithm and show that with the aid of RIS, the target detection performance can be effectively enhanced.

LDHD‐Net: A Lightweight Network With Double Branch Head for Feature Enhancement of UAV Targets in Complex Scenes

The development of small UAV technology has led to the emergence of new challenges in UAV countermeasures. The timely detection of UAVs can effectively prevent potential infringements on airspace and privacy. Currently, methods based on deep learning demonstrate excellent performance in target detection. However, in complex scenes, there is a tendency for false alarms (FAs) and misdetections to occur at a higher rate. To solve these problems, we propose a lightweight infrared small target detection algorithm LDHD‐Net. First, we design a novel Ghost‐Shuffle module in the backbone network to enhance the network feature extraction capability. Meanwhile, we remove redundant layers from the network to make the backbone network more lightweight. Second, we design a hierarchical attention enhancement module in the neck network to improve the saliency of UAV targets and reduce background noise interference. In addition, we design a novel small target detection structure and prediction heads in the shallow layers of the network to improve small target detection accuracy. Finally, we design a novel attention dual‐branch head to reduce interference between different tasks and improve the real‐time performance of algorithm detection. The experimental results show that compared with the original model, inference time remains essentially the same, LDHD‐Net parameters are only 3.9 M and AP improves by 12.6%. Compared to SOTA methods, LDHD‐Net shows better performance on SIDD and Anti‐UAV410 datasets. The algorithm effectively improves the accuracy and real‐time detection of UAVs in complex scenes.

Small target detection algorithm based on multi-branch stacking and new sampling transition module.

Aiming at the problem that the SSD algorithm does not fully extract the feature information contained in each feature layer, as well as the feature information is easily lost during the sampling process, which makes the feature expression ineffective and leads to insufficient performance in small target detection. In this paper, AMT-SSD is proposed, a small target detection algorithm that incorporates the multi-branch stacking and new sampling transition module of the attention mechanism. In this algorithm, the composite attention mechanism is utilized to improve the correlation of features of the samples to be detected in terms of spatial and channels, and the efficiency of the algorithm; secondly, multi-branch stacking module is used to extract multi-size features for each feature layer, and different sizes of convolution kernels are utilized in parallel to fully extract their features and improve the expression of features; meanwhile, during the sampling process, the problem of missing features is solved by applying inverse subpixel convolution in the new sampling transition module. Experimentally, the AMT-SSD algorithm achieves 84.6% and 53.4% mAP metrics on the PASCAL VOC dataset and MS COCO dataset, respectively. This indicates that the AMT-SSD algorithm can effectively extract feature information that is beneficial to detection samples, and also performs well in reducing feature loss, which is effective for the algorithm to improve the algorithm on small targets.

Analysis of Detection Performance of Target Detection by Passive Interferometric Microwave Sensors (PIMSs)

Analysis of Detection Performance of Target Detection by Passive Interferometric Microwave Sensors (PIMSs)

Evaluation of maneuvering target detection performance based on measured data

Evaluation of maneuvering target detection performance based on measured data

STD-YOLOv8: A lightweight small target detection algorithm for UAV perspectives

<p>When recognizing targets by unmanned aerial vehicles (UAVs), problems such as small size, dense dispersion, and complex background are likely to occur, resulting in low recognition rates. In order to solve the above problems, this work proposed a lightweight small target detection algorithm based on the YOLOv8n: STD-YOLOv8 algorithm. First, the regression problem of small targets in different training periods was optimized, the penalty term in the original loss was improved, and a new LIoU loss function was proposed, so that the size of the penalty term could be dynamically adjusted before and after training, thus improving the performance of the algorithm. Second, in order to better adapt to the small target scale and enhance the ability of small target feature acquisition, the SPD-Conv module was integrated in the backbone network, replacing the original stepwise convolutional layer and pooling layer, so as to solve the problems of loss of fine-grained information and low efficiency of feature representation existing in the current convolutional neural network (CNN) structure. In the neck part, nearest-neighbor upsampling was replaced by the feature reassembly assembly of features operator CARAFE (content-aware reassembly of features), which enabled the model to aggregate contextual information in a larger perceptual field and enhanced the feature representation in the neck. Finally, validation experiments were conducted by comparing different algorithms under the same VisDrone-2021 dataset. The results of the ablation experiments show that the algorithms proposed in this thesis have improved the recall (R), mAP50, and mAP95 by 4.7, 5.8 and 5.7%, respectively, compared with YOLOv8n. The results of the model generalization experiments on the TinyPerson dataset show that the algorithm in this paper has superior small target detection performance with only 1.2 M model parameters (1 M = 10<sup>6</sup>).</p>

Intelligent Gangue Sorting System Based on Dual-Energy X-ray and Improved YOLOv5 Algorithm

Intelligent gangue sorting with high precision is of vital importance for improving coal quality. To tackle the challenges associated with coal gangue target detection, including algorithm performance imbalance and hardware deployment difficulties, in this paper, an intelligent gangue separation system that adopts the elevated YOLO-v5 algorithm and dual-energy X-rays is proposed. Firstly, images of dual-energy X-ray transmission coal gangue mixture under the actual operation of a coal mine were collected, and datasets for training and validation were self-constructed. Then, in the YOLOv5 backbone network, the EfficientNetv2 was used to replace the original cross stage partial darknet (CSPDarknet) to achieve the lightweight of the backbone network; in the neck, a light path aggregation network (LPAN) was designed based on PAN, and a convolutional block attention module (CBAM) was integrated into the BottleneckCSP of the feature fusion block to raise the feature acquisition capability of the network and maximize the learning effect. Subsequently, to accelerate the rate of convergence, an efficient intersection over union (EIOU) was used instead of the complete intersection over union (CIOU) loss function. Finally, to address the problem of low resolution of small targets leading to missed detection, an L2 detection head was introduced to the head section to improve the multi-scale target detection performance of the algorithm. The experimental results indicate that in comparison with YOLOv5-S, the same version of the algorithm proposed in this paper increases by 19.2% and 32.4% on mAP @.5 and mAP @.5:.95, respectively. The number of parameters decline by 51.5%, and the calculation complexity declines by 14.7%. The algorithm suggested in this article offers new ideas for the design of identification algorithms for coal gangue sorting systems, which is expected to save energy and reduce consumption, reduce labor, improve efficiency, and be more friendly to the embedded platform.

Weed target detection at seedling stage in paddy fields based on YOLOX

Weeds are one of the greatest threats to the growth of rice, and the loss of crops is greater in the early stage of rice growth. Traditional large-area spraying cannot selectively spray weeds and can easily cause herbicide waste and environmental pollution. To realize the transformation from large-area spraying to precision spraying in rice fields, it is necessary to quickly and efficiently detect the distribution of weeds. Benefiting from the rapid development of vision technology and deep learning, this study applies a computer vision method based on deep-learning-driven rice field weed target detection. To address the need to identify small dense targets at the rice seedling stage in paddy fields, this study propose a method for weed target detection based on YOLOX, which is composed of a CSPDarknet backbone network, a feature pyramid network (FPN) enhanced feature extraction network and a YOLO Head detector. The CSPDarknet backbone network extracts feature layers with dimensions of 80 pixels ⊆ 80 pixels, 40 pixels ⊆ 40 pixels and 20 pixels ⊆ 20 pixels. The FPN fuses the features from these three scales, and YOLO Head realizes the regression of the object classification and prediction boxes. In performance comparisons of different models, including YOLOv3, YOLOv4-tiny, YOLOv5-s, SSD and several models of the YOLOX series, namely, YOLOX-s, YOLOX-m, YOLOX-nano, and YOLOX-tiny, the results show that the YOLOX-tiny model performs best. The mAP, F1, and recall values from the YOLOX-tiny model are 0.980, 0.95, and 0.983, respectively. Meanwhile, the intermediate variable memory generated during the model calculation of YOLOX-tiny is only 259.62 MB, making it suitable for deployment in intelligent agricultural devices. However, although the YOLOX-tiny model is the best on the dataset in this paper, this is not true in general. The experimental results suggest that the method proposed in this paper can improve the model performance for the small target detection of sheltered weeds and dense weeds at the rice seedling stage in paddy fields. A weed target detection model suitable for embedded computing platforms is obtained by comparing different single-stage target detection models, thereby laying a foundation for the realization of unmanned targeted herbicide spraying performed by agricultural robots.

Revolutionizing Target Detection in Intelligent Traffic Systems: YOLOv8-SnakeVision

Intelligent traffic systems represent one of the crucial domains in today’s world, aiming to enhance traffic management efficiency and road safety. However, current intelligent traffic systems still face various challenges, particularly in the realm of target detection. These challenges include adapting to complex traffic scenarios and the lack of precise detection for multiple objects. To address these issues, we propose an innovative approach known as YOLOv8-SnakeVision. This method introduces Dynamic Snake Convolution, Context Aggregation Attention Mechanisms, and the Wise-IoU strategy within the YOLOv8 framework to enhance target detection performance. Dynamic Snake Convolution assists in accurately capturing complex object shapes and features, especially in cases of target occlusion or overlap. The Context Aggregation Attention Mechanisms allow the model to better focus on critical image regions and effectively integrate information, thus improving its ability to recognize obscured targets, small objects, and complex patterns. The Wise-IoU strategy combines dynamic non-monotonic focusing mechanisms, aiming to more precisely regress target bounding boxes, particularly for low-quality examples. We validate our approach on the BDD100K and NEXET datasets. Experimental results demonstrate that YOLOv8-SnakeVision excels in various complex road traffic scenarios. It not only enhances small object detection but also strengthens the ability to recognize multiple targets. This innovative method provides robust support for the development of intelligent traffic systems and holds the promise of achieving further breakthroughs in future applications.

Weed Detection Method Based on Lightweight and Contextual Information Fusion

Weed detection technology is of paramount significance in achieving automation and intelligence in weed control. Nevertheless, it grapples with several formidable challenges, including imprecise small target detection, high computational demands, inadequate real-time performance, and susceptibility to environmental background interference. In response to these practical issues, we introduce CCCS-YOLO, a lightweight weed detection algorithm, built upon enhancements to the Yolov5s framework. In this study, the Faster_Block is integrated into the C3 module of the YOLOv5s neck network, creating the C3_Faster module. This modification not only streamlines the network but also significantly amplifies its detection capabilities. Subsequently, the context aggregation module is enhanced in the head by improving the convolution blocks, strengthening the network’s ability to distinguish between background and targets. Furthermore, the lightweight Content-Aware ReAssembly of Feature (CARAFE) module is employed to replace the upsampling module in the neck network, enhancing the performance of small target detection and promoting the fusion of contextual information. Finally, Soft-NMS-EIoU is utilized to replace the NMS and CIoU modules in YOLOv5s, enhancing the accuracy of target detection under dense conditions. Through detection on a publicly available sugar beet weed dataset and sesame weed datasets, the improved algorithm exhibits significant improvement in detection performance compared to YOLOv5s and demonstrates certain advancements over classical networks such as YOLOv7 and YOLOv8.

Spaceborne distributed aperture radar maneuvering target detection approach with space-time 2D hybrid integration technique

Moving target detection in spaceborne distributed aperture radar (SBDAR) is a challenging task in civilian and military fields. The common issues are that the current methods suffer from the large range cell migration (RCM) and time-varying Doppler frequency modulation (DFM) problems in space-time two-dimensional (2D) domain, which is caused by distributed aperture radar detection configuration and the maneuvering characteristics of moving targets. To address these issues, a novel SBDAR moving target detection approach via space–time hybrid integration (STHI) is proposed in this paper. Compared with single or bistatic radar mode, the utilization of multiple satellites provides an improvement in moving target detection performance, together with the capabilities of localization and parameters estimation. The proposed method mainly includes the following three steps. First, the geometric and signal models of an observed maneuvering target are accurately established in a SBDAR framework, where the space-time 2D RCM and DFM are analyzed. Based on that, the second-order Keystone transform (SKT), a novel range frequency reversal process (NRFRP) and a modified scaled Fourier transform (MSCFT) are developed to correct the RCM and DFM, achieving an efficient time dimension coherent integration for maneuvering target. After that, by dividing the radar detection area, the spatial projection method is adopted to integrate the moving target energy in space dimension. Finally, the simulation results verify the effectiveness and advantages of the proposed method.