Small Target Detection Performance Research Articles

Small Target Detection Algorithm Based on Improved YOLOv5

Small targets exist in large numbers in various fields. They are broadly used in aerospace, video monitoring, and industrial detection. However, because of its tiny dimensions and modest resolution, the precision of small-target detection is low, and the erroneous detection rate is high. Therefore, based on YOLOv5, an improved small-target detection model is proposed. First, in order to improve the number of tiny targets detected while enhancing small-target detection performance, an additional detection head is added. Second, involution is used between the backbone and neck to increase the channel information of feature mapping. Third, the model introduces the BiFormer, wherein both the global and local feature information are captured simultaneously by means of its double-layer routing attention mechanism. Finally, a context augmentation module (CAM) is inserted into the neck in order to maximize the structure of feature fusion. In addition, in order to consider among the required real frame as well as the prediction frame simultaneously, YOLOv5’s original loss function is exchanged. The experimental results using the public dataset VisDrone2019 show that the proposed model has P increased by 13.43%, R increased by 11.28%, and mAP@.5 and mAP@[.5:.95] increased by 13.88% and 9.01%, respectively.

High-Precision Instance Segmentation Detection of Micrometer-Scale Primary Carbonitrides in Nickel-Based Superalloys for Industrial Applications

In industrial production, the identification and characterization of micron-sized second phases, such as carbonitrides in alloys, hold significant importance for optimizing alloy compositions and processes. However, conventional methods based on threshold segmentation suffer from drawbacks, including low accuracy, inefficiency, and subjectivity. Addressing these limitations, this study introduced a carbonitride instance segmentation model tailored for various nickel-based superalloys. The model enhanced the YOLOv8n network structure by integrating the SPDConv module and the P2 small target detection layer, thereby augmenting feature fusion capability and small target detection performance. Experimental findings demonstrated notable improvements: the mAP50 (Box) value increased from 0.676 to 0.828, and the mAP50 (Mask) value from 0.471 to 0.644 for the enhanced YOLOv8n model. The proposed model for carbonitride detection surpassed traditional threshold segmentation methods, meeting requirements for precise, rapid, and batch-automated detection in industrial settings. Furthermore, to assess the carbonitride distribution homogeneity, a method for quantifying dispersion uniformity was proposed and integrated into a data processing framework for seamless automation from prediction to analysis.

An effective graph embedded YOLOv5 model for forest fire detection

AbstractThe existing YOLOv5‐based framework has achieved great success in the field of target detection. However, in forest fire detection tasks, there are few high‐quality forest fire images available, and the performance of the YOLO model has suffered a serious decline in detecting small‐scale forest fires. Making full use of context information can effectively improve the performance of small target detection. To this end, this paper proposes a new graph‐embedded YOLOv5 forest fire detection framework, which can improve the performance of small‐scale forest fire detection using different scales of context information. To mine local context information, we design a spatial graph convolution operation based on the message passing neural network (MPNN) mechanism. To utilize global context information, we introduce a multi‐head self‐attention (MSA) module before each YOLO head. The experimental results on FLAME and our self‐built fire dataset show that our proposed model improves the accuracy of small‐scale forest fire detection. The proposed model achieves high performance in real‐time performance by fully utilizing the advantages of the YOLOv5 framework.

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>

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.

Tomato Maturity Recognition Model Based on Improved YOLOv5 in Greenhouse

Due to the dense distribution of tomato fruit with similar morphologies and colors, it is difficult to recognize the maturity stages when the tomato fruit is harvested. In this study, a tomato maturity recognition model, YOLOv5s-tomato, is proposed based on improved YOLOv5 to recognize the four types of different tomato maturity stages: mature green, breaker, pink, and red. Tomato maturity datasets were established using tomato fruit images collected at different maturing stages in the greenhouse. The small-target detection performance of the model was improved by Mosaic data enhancement. Focus and Cross Stage Partial Network (CSPNet) were adopted to improve the speed of network training and reasoning. The Efficient IoU (EIoU) loss was used to replace the Complete IoU (CIoU) loss to optimize the regression process of the prediction box. Finally, the improved algorithm was compared with the original YOLOv5 algorithm on the tomato maturity dataset. The experiment results show that the YOLOv5s-tomato reaches a precision of 95.58% and the mean Average Precision (mAP) is 97.42%; they are improved by 0.11% and 0.66%, respectively, compared with the original YOLOv5s model. The per-image detection speed is 9.2 ms, and the size is 23.9 MB. The proposed YOLOv5s-tomato can effectively solve the problem of low recognition accuracy for occluded and small-target tomatoes, and it also can meet the accuracy and speed requirements of tomato maturity recognition in greenhouses, making it suitable for deployment on mobile agricultural devices to provide technical support for the precise operation of tomato-picking machines.

YOLOv4-A: Research on Traffic Sign Detection Based on Hybrid Attention Mechanism

<p>Aiming at the problem of false detection and missed detection in the traffic sign detection task, an improved YOLOv4 detection algorithm is proposed. Based on the YOLOv4 algorithm, the Efficient Channel Attention Module (ECA) and the Convolutional Block Attention Module (CBAM) are added to form YOLOv4-A algorithm. At the same time, the global K-means clustering algorithm is used to regenerate smaller anchors, which makes the network converge faster and reduces the error rate. The YOLOv4-A algorithm re-calibrates the detection branch features in the two dimensions of channel and space, so that the network can focus and enhance the effective features, and suppress the interference features, which improves the detection ability of the algorithm. Experiments on the TT100K traffic sign dataset show that the proposed algorithm has a particularly significant improvement in the performance of small target detection. Compared with the YOLOv4 algorithm, the precision and mAP@0.5 of the proposed algorithm are increased by 5.38% and 5.75%.</p> <p> </p>

PointTransformer: Encoding Human Local Features for Small Target Detection.

The improvement of small target detection and obscuration handling is the key problem to be solved in the object detection task. In the field operation of chemical plant, due to the occlusion of construction workers and the long distance of surveillance shooting, it often leads to the phenomenon of missed detection. Most of the existing work uses multiple feature fusion strategies to extract different levels of features and then aggregate them into global features, which does not utilize local features and makes it difficult to improve the performance of small target detection. To address this issue, this paper introduces Point Transformer, a transformer encoder, as the core backbone of the object detection framework that first uses a priori information of human skeletal points to obtain local features and then uses both self-attention and cross-attention mechanisms to reconstruct the local features corresponding to each key point. In addition, since the target to be detected is highly correlated with the position of human skeletal points, to further boost Point Transformer's performance, a learnable positional encoding method is proposed by us to highlight the position characteristics of each skeletal point. The proposed model is evaluated on the dataset of field operation in a chemical plant. The results are significantly better than the classical algorithms. It also outperforms state-of-the-art by 12 percent of map points in the small target detection task.

Faster object detector for drone-captured images

Object detection based on computer vision is becoming popular in drone-captured images. However, real-time object detection in unmanned aerial vehicle (UAV) scenarios is a huge challenge for low-end devices. To deal with the problem, we have improved YOLOv3-tiny in the following aspects. First, the label rewriting problem, which is caused by network structure and dataset of YOLOv3-tiny in drone-captured images detection, is very serious. The method of increasing the size of the predicted feature map is used to reduce the ratio of label rewriting. Second, the features of small targets will be reduced in a small feature map, but the context information with large receptive fields in it can improve the performance of small target detection. So we use dilated convolution to expand the receptive field without reducing the size of the feature map. Third, multiscale feature fusion is very helpful for small target detection. The multidilated module is adopted to merge features in earlier layer and deeper layers. Finally, a pretraining strategy combining copy-paste data augmentation method is proposed to learn more features from categories with a small number of samples. We evaluated our model on the VisDrone2019-Det test set. It achieves compelling results compared to the counterparts of YOLOv3-tiny, including ∼86.1 % decline in model size, increasing ∼19.2 % AP50. Although our model is slower than YOLOv3-tiny, it is 2.96 times faster than YOLOv3. The results of experiments verify that our network is more effective than YOLOv3-tiny. It is more suitable for UAV object detection applications on low-end devices.

Double layer local contrast measure and multi-directional gradient comparison for small infrared target detection

Infrared small target detection is one of the key technologies in the search and track (IRST) based on infrared imaging equipment. At present, the performance of small target detection based on single frame infrared image is directly related to the accuracy of subsequent target tracking, so it has been studied a lot. However, the existing small target detection algorithms have certain limitations in detection accuracy and real-time performance, especially when the contrast between the target and the background area is not high or the background is complex, especially in the complex sea or sky background, due to the influence of a large amount of noise and clutter in the background, the existing infrared small target detection algorithms have a high false alarm rate. To solve the above problems, this paper proposes a small target detection algorithm based on weighted double layer local contrast and multi-directional gradient map, which realizes the accurate detection of small targets from two aspects of targets’ local contrast and gradient. Firstly, we design an improved two layer local contrast measurement architecture, and use the weighted mean method to better represent the gray value of the local window; Secondly, a local contrast comparison method based on target and background is proposed to enhance the intensity of small targets and suppress some background clutter; Then, the multi-directional gradient map is used to further suppress the noise so as to improve the contrast between the target and the background. At the same time, singular value decomposition (SVD) method is used to extract the main features including small targets, which can effectively suppress the small texture interference around the targets in the background without losing the target intensity; Finally, an adaptive threshold method is used to separate small targets from their background. Experimental results show that compared with the existing algorithms, the proposed detection algorithm can effectively reduce the false alarm rate in different complex scenes, and the computational efficiency is improved compared with some multi-scale small target detection methods. At the same time, the signal to clutter ratio (SCR), background suppression factor (BSF) and receiver operating characteristic (ROC) curve are also better than these existing state of the art algorithms, which can display good robustness.

Large-Difference-Scale Target Detection Using a Revised Bhattacharyya Distance in SAR Images

Small target detection is a very challenging problem since a small target contains only a few pixels in size. At present, many deep learning-based detection algorithms for small targets have achieved remarkable results, mainly including improvements in data augmentation, multiscale images, multiscale features, training strategies, and so on. However, these deep learning-based methods cannot select the positive and negative samples for the large-difference-scale targets well in the label assignment operation. The reason is that intersection over union (IoU), which is widely used in the most target detection networks, has great limitations for small target detection. However, in practical applications, there are often some large-difference-scale targets in synthetic aperture radar (SAR) images, especially existing some tiny targets due to the limitation of resolution. To fundamentally break the limitations of IoU, we propose to use the Bhattacharyya distance (BD) instead of the IoU metric to improve the performance of small target detection. We further revise the Bhattacharyya distance (RBD) to better measure the deviation of bounding boxes for targets with large differences in size. RBD can embed anchor-based detectors to replace the IoU metric in label assignment and nonmaximum suppression (NMS). The proposed method is evaluated on the LS-SSDD-v1.0 dataset and the experimental results show that the proposed method outperforms the state-of-the-art methods.

Infrared Small Target Detection Based on Weighted Three-Layer Window Local Contrast

The performance of small target detection restricts the development of the infrared search and track (IRST) system. Against the complicated background clutter of the infrared (IR) image, the small targets are difficult to separate from a noisy background. Aiming at solving the problem of residual background clutter in the local contrast method, a weighted three-layer window local contrast method (WTLLCM) is proposed in this letter. First, the images are filtered by a layered gradient kernel to enhance the contrast between targets and background. Then, a three-layer window is utilized to calculate the local contrast of the filtered images. Next, it is worth mentioning that a simple target aggregation strategy is considered to preserve the integrity of the target. Especially, a new region intensity level (NRIL) algorithm is proposed to weigh the local contrast map to further suppress the background. Finally, the targets are detected by adaptive threshold segmentation. Compared with state-of-the-art small targets detection baseline algorithms based on local contrast, extensive experimental results demonstrate the superiority of the proposed method, especially in complex backgrounds. And instead of utilizing multi-scale windows, multi-scale targets detection is accomplished by using a single-scale window to reduce the calculation of the method.

ECAP-YOLO: Efficient Channel Attention Pyramid YOLO for Small Object Detection in Aerial Image

Detection of small targets in aerial images is still a difficult problem due to the low resolution and background-like targets. With the recent development of object detection technology, efficient and high-performance detector techniques have been developed. Among them, the YOLO series is a representative method of object detection that is light and has good performance. In this paper, we propose a method to improve the performance of small target detection in aerial images by modifying YOLOv5. The backbone is was modified by applying the first efficient channel attention module, and the channel attention pyramid method was proposed. We propose an efficient channel attention pyramid YOLO (ECAP-YOLO). Second, in order to optimize the detection of small objects, we eliminated the module for detecting large objects and added a detect layer to find smaller objects, reducing the computing power used for detecting small targets and improving the detection rate. Finally, we use transposed convolution instead of upsampling. Comparing the method proposed in this paper to the original YOLOv5, the performance improvement for the mAP was 6.9% when using the VEDAI dataset, 5.4% when detecting small cars in the xView dataset, 2.7% when detecting small vehicle and small ship classes from the DOTA dataset, and approximately 2.4% when finding small cars in the Arirang dataset.

Remote Sensing Data Detection Based on Multiscale Fusion and Attention Mechanism

Remote sensing images are often of low quality due to the limitations of the equipment, resulting in poor image accuracy, and it is extremely difficult to identify the target object when it is blurred or small. The main challenge is that objects in sensing images have very few pixels. Traditional convolutional networks are complicated to extract enough information through local convolution and are easily disturbed by noise points, so they are usually not ideal for classifying and diagnosing small targets. The current solution is to process the feature map information at multiple scales, but this method does not consider the supplementary effect of the context information of the feature map on the semantics. In this work, in order to enable CNNs to make full use of context information and improve its representation ability, we propose a residual attention function fusion method, which improves the representation ability of feature maps by fusing contextual feature map information of different scales, and then propose a spatial attention mechanism for global pixel point convolution response. This method compresses global pixels through convolution, weights the original feature map pixels, reduces noise interference, and improves the network’s ability to grasp global critical pixel information. In experiments, the remote sensing ship image recognition experiments on remote sensing image data sets show that the network structure can improve the performance of small-target detection. The results on cifar10 and cifar100 prove that the attention mechanism is universal and practical.

Remote sensing target detection in a harbor area based on an arbitrary-oriented convolutional neural network

With the rapid development of optical remote sensing, it is urgent to find a reliable target detection method. Compared with traditional detection algorithms, a convolutional neural network has attracted considerable attention owing to its efficiency and high transitivity. However, different from general images, remote sensing images contain complex background information and dense small targets with changeable directions that make detection very challenging. To solve these problems and provide a stable and high-performance detection method, a rotated saliency fusion object detection (RSD) model based on “you only look once” (YOLO)v4 is established. First, salient image fusion technology is used to magnify target information. Second, the angle variable and rotated non-maximal suppression is introduced to improve the accuracy of rotated object detection by including the detection of dense objects. Third, the network structure is enhanced to improve the performance of small-target detection. Finally, the k-means algorithm and data enhancement are introduced to increase the robustness of the model. Extensive experiments demonstrate the superiority of the proposed model in detection speed and accuracy. The mean average precision of the proposed RSD model reaches 97.32% for the remote sensing images in a harbor area with an average detection speed of 13.41 s − 1.

A Novel SAR Image Ship Small Targets Detection Method

To satisfy practical requirements of high real-time accuracy and low computational complexity of synthetic aperture radar (SAR) image ship small target detection, this paper proposes a small ship target detection method based on the improved You Only Look Once Version 3 (YOLOv3). The main contributions of this study are threefold. First, the feature extraction network of the original YOLOV3 algorithm is replaced with the VGG16 network convolution layer. Second, general convolution is transformed into depthwise separable convolution, thereby reducing the computational cost of the algorithm. Third, a residual network structure is introduced into the feature extraction network to reuse the shallow target feature information, which enhances the detailed features of the target and ensures the improvement in accuracy of small target detection performance. To evaluate the performance of the proposed method, many experiments are conducted on public SAR image datasets. For ship targets with complex backgrounds and small ship targets in the SAR image, the effectiveness of the proposed algorithm is verified. Results show that the accuracy and recall rate improved by 5.31% and 2.77%, respectively, compared with the original YOLOV3. Furthermore, the proposed model not only significantly reduces the computational effort, but also improves the detection accuracy of ship small target.

Enhancing Small Moving Target Detection Performance in Low-Quality and Long-Range Infrared Videos Using Optical Flow Techniques

The detection of small moving objects in long-range infrared videos is challenging due to background clutter, air turbulence, and small target size. In this paper, we summarize the investigation of efficient ways to enhance the performance of small target detection in long-range and low-quality infrared videos containing moving objects. In particular, we focus on unsupervised, modular, flexible, and efficient methods for target detection performance enhancement using motion information extracted from optical flow methods. Three well-known optical flow methods were studied. It was found that optical flow methods need to be combined with contrast enhancement, connected component analysis, and target association in order to be effective for target detection. Extensive experiments using long-range mid-wave infrared (MWIR) videos from the Defense Systems Information Analysis Center (DSIAC) dataset clearly demonstrated the efficacy of our proposed approach.

Infrared Small Target Detection based on Principal Component Tracing

According to the characteristics of infrared small target images, an infrared small target detection method based on principal component tracing is proposed. The method by minimizing the combination of the nuclear matrix norm and L1 norm implementation principal component tracing, using principal component tracing to the infrared small target image is decomposed into sparse matrix and low rank matrix, sparse matrix for segmented small target detection. The proposed method is tested and compared with the existing method. The experimental results show that the proposed infrared small target detection method has better small target detection performance.

Infrared small target detection in heavy sky scene clutter based on sparse representation

A novel infrared small target detection method based on sky clutter and target sparse representation is proposed in this paper to cope with the representing uncertainty of clutter and target. The sky scene background clutter is described by fractal random field, and it is perceived and eliminated via the sparse representation on fractal background over-complete dictionary (FBOD). The infrared small target signal is simulated by generalized Gaussian intensity model, and it is expressed by the generalized Gaussian target over-complete dictionary (GGTOD), which could describe small target more efficiently than traditional structured dictionaries. Infrared image is decomposed on the union of FBOD and GGTOD, and the sparse representation energy that target signal and background clutter decomposed on GGTOD differ so distinctly that it is adopted to distinguish target from clutter. Some experiments are induced and the experimental results show that the proposed approach could improve the small target detection performance especially under heavy clutter for background clutter could be efficiently perceived and suppressed by FBOD and the changing target could also be represented accurately by GGTOD.

Adaptive top-hat filter based on quantum genetic algorithm for infrared small target detection

With the development of infrared technology, infrared small targets detection has attracted great interest of researchers. Top-hat filter is one of widely used methods for detecting infrared small target, and the structure elements have great influence on the performance of detection. The structure elements are desired to be adjusted adaptively. To this end, an adaptive structure elements optimization method based on quantum genetic algorithm (QGA) is introduced, and the convergence of QGA reveals the effectiveness of QGA. Experimental results show that the proposed adaptive top-hat filter based on QGA can achieve more stable infrared small target detection performance compared with the traditional top-hat filter.