基于改进Detection Transformer的棉花幼苗与杂草检测模型研究

基于深度学习的目标检测技术在棉花幼苗与杂草检测领域已取得一定进展。论文提出了基于改进Detection Transformer的棉花幼苗与杂草检测模型,以提高杂草目标检测的准确率和效率。首先,引入了可变形注意力模块替代原始模型中的Transformer注意力模块,提高模型对特征图目标形变的处理能力。提出新的降噪训练机制,解决了二分图匹配不稳定问题。提出混合查询选择策略,提高解码器对目标类别和位置信息的利用效率。使用Swin Transformer作为网络主干,提高模型特征提取能力。通过对比原网络,论文提出的模型方法在训练过程中表现出更快的收敛速度,并且在准确率方面提高了6.7%。

Esophageal cancer screening,early detection and treatment:Current insights and future directions

Esophageal cancer ranks among the most prevalent malignant tumors globally,primarily due to its highly aggressive nature and poor survival rates.According to the 2020 global cancer statistics,there were approximately 604000 new cases of esophageal cancer,resulting in 544000 deaths.The 5-year survival rate hovers around a mere 15%-25%.Notably,distinct variations exist in the risk factors associated with the two primary histological types,influencing their worldwide incidence and distribution.Squamous cell carcinoma displays a high incidence in specific regions,such as certain areas in China,where it meets the cost-effect-iveness criteria for widespread endoscopy-based early diagnosis within the local population.Conversely,adenocarcinoma(EAC)represents the most common histological subtype of esophageal cancer in Europe and the United States.The role of early diagnosis in cases of EAC originating from Barrett's esophagus(BE)remains a subject of controversy.The effectiveness of early detection for EAC,particularly those arising from BE,continues to be a debated topic.The variations in how early-stage esophageal carcinoma is treated in different regions are largely due to the differing rates of early-stage cancer diagnoses.In areas with higher incidences,such as China and Japan,early diagnosis is more common,which has led to the advancement of endoscopic methods as definitive treatments.These techniques have demonstrated remarkable efficacy with minimal complications while preserving esophageal functionality.Early screening,prompt diagnosis,and timely treatment are key strategies that can significantly lower both the occurrence and death rates associated with esophageal cancer.

Detection of Turbulence Anomalies Using a Symbolic Classifier Algorithm in Airborne Quick Access Record(QAR)Data Analysis

As the risks associated with air turbulence are intensified by climate change and the growth of the aviation industry,it has become imperative to monitor and mitigate these threats to ensure civil aviation safety.The eddy dissipation rate(EDR)has been established as the standard metric for quantifying turbulence in civil aviation.This study aims to explore a universally applicable symbolic classification approach based on genetic programming to detect turbulence anomalies using quick access recorder(QAR)data.The detection of atmospheric turbulence is approached as an anomaly detection problem.Comparative evaluations demonstrate that this approach performs on par with direct EDR calculation methods in identifying turbulence events.Moreover,comparisons with alternative machine learning techniques indicate that the proposed technique is the optimal methodology currently available.In summary,the use of symbolic classification via genetic programming enables accurate turbulence detection from QAR data,comparable to that with established EDR approaches and surpassing that achieved with machine learning algorithms.This finding highlights the potential of integrating symbolic classifiers into turbulence monitoring systems to enhance civil aviation safety amidst rising environmental and operational hazards.

Defect Detection Model Using Time Series Data Augmentation and Transformation

Time-series data provide important information in many fields,and their processing and analysis have been the focus of much research.However,detecting anomalies is very difficult due to data imbalance,temporal dependence,and noise.Therefore,methodologies for data augmentation and conversion of time series data into images for analysis have been studied.This paper proposes a fault detection model that uses time series data augmentation and transformation to address the problems of data imbalance,temporal dependence,and robustness to noise.The method of data augmentation is set as the addition of noise.It involves adding Gaussian noise,with the noise level set to 0.002,to maximize the generalization performance of the model.In addition,we use the Markov Transition Field(MTF)method to effectively visualize the dynamic transitions of the data while converting the time series data into images.It enables the identification of patterns in time series data and assists in capturing the sequential dependencies of the data.For anomaly detection,the PatchCore model is applied to show excellent performance,and the detected anomaly areas are represented as heat maps.It allows for the detection of anomalies,and by applying an anomaly map to the original image,it is possible to capture the areas where anomalies occur.The performance evaluation shows that both F1-score and Accuracy are high when time series data is converted to images.Additionally,when processed as images rather than as time series data,there was a significant reduction in both the size of the data and the training time.The proposed method can provide an important springboard for research in the field of anomaly detection using time series data.Besides,it helps solve problems such as analyzing complex patterns in data lightweight.

Multimodal Social Media Fake News Detection Based on Similarity Inference and Adversarial Networks

As social networks become increasingly complex, contemporary fake news often includes textual descriptionsof events accompanied by corresponding images or videos. Fake news in multiple modalities is more likely tocreate a misleading perception among users. While early research primarily focused on text-based features forfake news detection mechanisms, there has been relatively limited exploration of learning shared representationsin multimodal (text and visual) contexts. To address these limitations, this paper introduces a multimodal modelfor detecting fake news, which relies on similarity reasoning and adversarial networks. The model employsBidirectional Encoder Representation from Transformers (BERT) and Text Convolutional Neural Network (Text-CNN) for extracting textual features while utilizing the pre-trained Visual Geometry Group 19-layer (VGG-19) toextract visual features. Subsequently, the model establishes similarity representations between the textual featuresextracted by Text-CNN and visual features through similarity learning and reasoning. Finally, these features arefused to enhance the accuracy of fake news detection, and adversarial networks have been employed to investigatethe relationship between fake news and events. This paper validates the proposed model using publicly availablemultimodal datasets from Weibo and Twitter. Experimental results demonstrate that our proposed approachachieves superior performance on Twitter, with an accuracy of 86%, surpassing traditional unimodalmodalmodelsand existing multimodal models. In contrast, the overall better performance of our model on the Weibo datasetsurpasses the benchmark models across multiple metrics. The application of similarity reasoning and adversarialnetworks in multimodal fake news detection significantly enhances detection effectiveness in this paper. However,current research is limited to the fusion of only text and image modalities. Future research directions should aimto further integrate features fromadditionalmodalities to comprehensively represent themultifaceted informationof fake news.

An Underwater Target Detection Algorithm Based on Attention Mechanism and Improved YOLOv7

For underwater robots in the process of performing target detection tasks,the color distortion and the uneven quality of underwater images lead to great difficulties in the feature extraction process of the model,which is prone to issues like error detection,omission detection,and poor accuracy.Therefore,this paper proposed the CER-YOLOv7(CBAM-EIOU-RepVGG-YOLOv7)underwater target detection algorithm.To improve the algorithm’s capability to retain valid features from both spatial and channel perspectives during the feature extraction phase,we have added a Convolutional Block Attention Module(CBAM)to the backbone network.The Reparameterization Visual Geometry Group(RepVGG)module is inserted into the backbone to improve the training and inference capabilities.The Efficient Intersection over Union(EIoU)loss is also used as the localization loss function,which reduces the error detection rate and missed detection rate of the algorithm.The experimental results of the CER-YOLOv7 algorithm on the UPRC(Underwater Robot Prototype Competition)dataset show that the mAP(mean Average Precision)score of the algorithm is 86.1%,which is a 2.2%improvement compared to the YOLOv7.The feasibility and validity of the CER-YOLOv7 are proved through ablation and comparison experiments,and it is more suitable for underwater target detection.

A core-satellite self-assembled SERS aptasensor containing a“biological-silent region”Raman tag for the accurate and ultrasensitive detection of histamine

Herein,a novel interference-free surface-enhanced Raman spectroscopy(SERS)strategy based on magnetic nanoparticles(MNPs)and aptamer-driven assemblies was proposed for the ultrasensitive detection of histamine.A core-satellite SERS aptasensor was constructed by combining aptamer-decorated Fe_(3)O_(4)@Au MNPs(as the recognize probe for histamine)and complementary DNA-modified silver nanoparticles carrying 4-mercaptobenzonitrile(4-MBN)(Ag@4-MBN@Ag-c-DNA)as the SERS signal probe for the indirect detection of histamine.Under an applied magnetic field in the absence of histamine,the assembly gave an intense Raman signal at“Raman biological-silent”region due to 4-MBN.In the presence of histamine,the Ag@4-MBN@Ag-c-DNA SERS-tag was released from the Fe_(3)O_(4)@Au MNPs,thus decreasing the SERS signal.Under optimal conditions,an ultra-low limit of detection of 0.65×10^(-3)ng/mL and a linear range 10^(-2)-10^5 ng/mL on the SERS aptasensor were obtained.The histamine content in four food samples were analyzed using the SERS aptasensor,with the results consistent with those determined by high performance liquid chromatography.The present work highlights the merits of indirect strategies for the ultrasensitive and highly selective SERS detection of small biological molecules in complex matrices.

Real-time Rescue Target Detection Based on UAV Imagery for Flood Emergency Response

Timely acquisition of rescue target information is critical for emergency response after a flood disaster.Unmanned Aerial Vehicles(UAVs)equipped with remote sensing capabilities offer distinct advantages,including high-resolution imagery and exceptional mobility,making them well suited for monitoring flood extent and identifying rescue targets during floods.However,there are some challenges in interpreting rescue information in real time from flood images captured by UAVs,such as the complexity of the scenarios of UAV images,the lack of flood rescue target detection datasets and the limited real-time processing capabilities of the airborne on-board platform.Thus,we propose a real-time rescue target detection method for UAVs that is capable of efficiently delineating flood extent and identifying rescue targets(i.e.,pedestrians and vehicles trapped by floods).The proposed method achieves real-time rescue information extraction for UAV platforms by lightweight processing and fusion of flood extent extraction model and target detection model.The flood inundation range is extracted by the proposed method in real time and detects targets such as people and vehicles to be rescued based on this layer.Our experimental results demonstrate that the Intersection over Union(IoU)for flood water extraction reaches an impressive 80%,and the IoU for real-time flood water extraction stands at a commendable 76.4%.The information on flood stricken targets extracted by this method in real time can be used for flood emergency rescue.

Automatic detection of small bowel lesions with different bleeding risks based on deep learning models

BACKGROUND Deep learning provides an efficient automatic image recognition method for small bowel(SB)capsule endoscopy(CE)that can assist physicians in diagnosis.However,the existing deep learning models present some unresolved challenges.AIM To propose a novel and effective classification and detection model to automatically identify various SB lesions and their bleeding risks,and label the lesions accurately so as to enhance the diagnostic efficiency of physicians and the ability to identify high-risk bleeding groups.METHODS The proposed model represents a two-stage method that combined image classification with object detection.First,we utilized the improved ResNet-50 classification model to classify endoscopic images into SB lesion images,normal SB mucosa images,and invalid images.Then,the improved YOLO-V5 detection model was utilized to detect the type of lesion and its risk of bleeding,and the location of the lesion was marked.We constructed training and testing sets and compared model-assisted reading with physician reading.RESULTS The accuracy of the model constructed in this study reached 98.96%,which was higher than the accuracy of other systems using only a single module.The sensitivity,specificity,and accuracy of the model-assisted reading detection of all images were 99.17%,99.92%,and 99.86%,which were significantly higher than those of the endoscopists’diagnoses.The image processing time of the model was 48 ms/image,and the image processing time of the physicians was 0.40±0.24 s/image(P<0.001).CONCLUSION The deep learning model of image classification combined with object detection exhibits a satisfactory diagnostic effect on a variety of SB lesions and their bleeding risks in CE images,which enhances the diagnostic efficiency of physicians and improves the ability of physicians to identify high-risk bleeding groups.

Human intrusion detection for high-speed railway perimeter under all-weather condition

Purpose – The paper aims to solve the problem of personnel intrusion identification within the limits of highspeed railways. It adopts the fusion method of millimeter wave radar and camera to improve the accuracy ofobject recognition in dark and harsh weather conditions.Design/methodology/approach – This paper adopts the fusion strategy of radar and camera linkage toachieve focus amplification of long-distance targets and solves the problem of low illumination by laser lightfilling of the focus point. In order to improve the recognition effect, this paper adopts the YOLOv8 algorithm formulti-scale target recognition. In addition, for the image distortion caused by bad weather, this paper proposesa linkage and tracking fusion strategy to output the correct alarm results.Findings – Simulated intrusion tests show that the proposed method can effectively detect human intrusionwithin 0–200 m during the day and night in sunny weather and can achieve more than 80% recognitionaccuracy for extreme severe weather conditions.Originality/value – (1) The authors propose a personnel intrusion monitoring scheme based on the fusion ofmillimeter wave radar and camera, achieving all-weather intrusion monitoring;(2) The authors propose a newmulti-level fusion algorithm based on linkage and tracking to achieve intrusion target monitoring underadverse weather conditions;(3) The authors have conducted a large number of innovative simulationexperiments to verify the effectiveness of the method proposed in this article.

A dual-RPA based lateral flow strip for sensitive,on-site detection of CP4-EPSPS and Cry1Ab/Ac genes in genetically modified crops

Traditional transgenic detection methods require high test conditions and struggle to be both sensitive and efficient.In this study,a one-tube dual recombinase polymerase amplification(RPA)reaction system for CP4-EPSPS and Cry1Ab/Ac was proposed and combined with a lateral flow immunochromatographic assay,named“Dual-RPA-LFD”,to visualize the dual detection of genetically modified(GM)crops.In which,the herbicide tolerance gene CP4-EPSPS and the insect resistance gene Cry1Ab/Ac were selected as targets taking into account the current status of the most widespread application of insect resistance and herbicide tolerance traits and their stacked traits.Gradient diluted plasmids,transgenic standards,and actual samples were used as templates to conduct sensitivity,specificity,and practicality assays,respectively.The constructed method achieved the visual detection of plasmid at levels as low as 100 copies,demonstrating its high sensitivity.In addition,good applicability to transgenic samples was observed,with no cross-interference between two test lines and no influence from other genes.In conclusion,this strategy achieved the expected purpose of simultaneous detection of the two popular targets in GM crops within 20 min at 37°C in a rapid,equipmentfree field manner,providing a new alternative for rapid screening for transgenic assays in the field.

Detection method of forward-scatter signal based on Rényi entropy

The application scope of the forward scatter radar(FSR)based on the Global Navigation Satellite System(GNSS)can be expanded by improving the detection capability.Firstly,the forward-scatter signal model when the target crosses the baseline is constructed.Then,the detection method of the for-ward-scatter signal based on the Rényi entropy of time-fre-quency distribution is proposed and the detection performance with different time-frequency distributions is compared.Simula-tion results show that the method based on the smooth pseudo Wigner-Ville distribution(SPWVD)can achieve the best perfor-mance.Next,combined with the geometry of FSR,the influence on detection performance of the relative distance between the target and the baseline is analyzed.Finally,the proposed method is validated by the anechoic chamber measurements and the results show that the detection ability has a 10 dB improvement compared with the common constant false alarm rate(CFAR)detection.

Confusing Object Detection:A Survey

Confusing object detection(COD),such as glass,mirrors,and camouflaged objects,represents a burgeoning visual detection task centered on pinpointing and distinguishing concealed targets within intricate backgrounds,leveraging deep learning methodologies.Despite garnering increasing attention in computer vision,the focus of most existing works leans toward formulating task-specific solutions rather than delving into in-depth analyses of methodological structures.As of now,there is a notable absence of a comprehensive systematic review that focuses on recently proposed deep learning-based models for these specific tasks.To fill this gap,our study presents a pioneering review that covers both themodels and the publicly available benchmark datasets,while also identifying potential directions for future research in this field.The current dataset primarily focuses on single confusing object detection at the image level,with some studies extending to video-level data.We conduct an in-depth analysis of deep learning architectures,revealing that the current state-of-the-art(SOTA)COD methods demonstrate promising performance in single object detection.We also compile and provide detailed descriptions ofwidely used datasets relevant to these detection tasks.Our endeavor extends to discussing the limitations observed in current methodologies,alongside proposed solutions aimed at enhancing detection accuracy.Additionally,we deliberate on relevant applications and outline future research trajectories,aiming to catalyze advancements in the field of glass,mirror,and camouflaged object detection.

Target Detection Algorithm in Foggy Scenes Based on Dual Subnets

Under the influence of air humidity,dust,aerosols,etc.,in real scenes,haze presents an uneven state.In this way,the image quality and contrast will decrease.In this case,It is difficult to detect the target in the image by the universal detection network.Thus,a dual subnet based on multi-task collaborative training(DSMCT)is proposed in this paper.Firstly,in the training phase,the Gated Context Aggregation Network(GCANet)is used as the supervisory network of YOLOX to promote the extraction of clean information in foggy scenes.In the test phase,only the YOLOX branch needs to be activated to ensure the detection speed of the model.Secondly,the deformable convolution module is used to improve GCANet to enhance the model’s ability to capture details of non-homogeneous fog.Finally,the Coordinate Attention mechanism is introduced into the Vision Transformer and the backbone network of YOLOX is redesigned.In this way,the feature extraction ability of the network for deep-level information can be enhanced.The experimental results on artificial fog data set FOG_VOC and real fog data set RTTS show that the map value of DSMCT reached 86.56%and 62.39%,respectively,which was 2.27%and 4.41%higher than the current most advanced detection model.The DSMCT network has high practicality and effectiveness for target detection in real foggy scenes.

YOLO-MFD:Remote Sensing Image Object Detection with Multi-Scale Fusion Dynamic Head

Remote sensing imagery,due to its high altitude,presents inherent challenges characterized by multiple scales,limited target areas,and intricate backgrounds.These inherent traits often lead to increased miss and false detection rates when applying object recognition algorithms tailored for remote sensing imagery.Additionally,these complexities contribute to inaccuracies in target localization and hinder precise target categorization.This paper addresses these challenges by proposing a solution:The YOLO-MFD model(YOLO-MFD:Remote Sensing Image Object Detection withMulti-scale Fusion Dynamic Head).Before presenting our method,we delve into the prevalent issues faced in remote sensing imagery analysis.Specifically,we emphasize the struggles of existing object recognition algorithms in comprehensively capturing critical image features amidst varying scales and complex backgrounds.To resolve these issues,we introduce a novel approach.First,we propose the implementation of a lightweight multi-scale module called CEF.This module significantly improves the model’s ability to comprehensively capture important image features by merging multi-scale feature information.It effectively addresses the issues of missed detection and mistaken alarms that are common in remote sensing imagery.Second,an additional layer of small target detection heads is added,and a residual link is established with the higher-level feature extraction module in the backbone section.This allows the model to incorporate shallower information,significantly improving the accuracy of target localization in remotely sensed images.Finally,a dynamic head attentionmechanism is introduced.This allows themodel to exhibit greater flexibility and accuracy in recognizing shapes and targets of different sizes.Consequently,the precision of object detection is significantly improved.The trial results show that the YOLO-MFD model shows improvements of 6.3%,3.5%,and 2.5%over the original YOLOv8 model in Precision,map@0.5 and map@0.5:0.95,separately.These results illustrate the clear advantages of the method.

Analysis of the joint detection capability of the SMILE satellite and EISCAT-3D radar

The Solar wind Magnetosphere Ionosphere Link Explorer(SMILE)satellite is a small magnetosphere–ionosphere link explorer developed cooperatively between China and Europe.It pioneers the use of X-ray imaging technology to perform large-scale imaging of the Earth’s magnetosheath and polar cusp regions.It uses a high-precision ultraviolet imager to image the overall configuration of the aurora and monitor changes in the source of solar wind in real time,using in situ detection instruments to improve human understanding of the relationship between solar activity and changes in the Earth’s magnetic field.The SMILE satellite is scheduled to launch in 2025.The European Incoherent Scatter Sciences Association(EISCAT)-3D radar is a new generation of European incoherent scatter radar constructed by EISCAT and is the most advanced ground-based ionospheric experimental device in the high-latitude polar region.It has multibeam and multidirectional quasi-real-time three-dimensional(3D)imaging capabilities,continuous monitoring and operation capabilities,and multiple-baseline interferometry capabilities.Joint detection by the SMILE satellite and the EISCAT-3D radar is of great significance for revealing the coupling process of the solar wind–magnetosphere–ionosphere.Therefore,we performed an analysis of the joint detection capability of the SMILE satellite and EISCAT-3D,analyzed the period during which the two can perform joint detection,and defined the key scientific problems that can be solved by joint detection.In addition,we developed Web-based software to search for and visualize the joint detection period of the SMILE satellite and EISCAT-3D radar,which lays the foundation for subsequent joint detection experiments and scientific research.

A Hybrid Feature Fusion Traffic Sign Detection Algorithm Based on YOLOv7

Autonomous driving technology has entered a period of rapid development,and traffic sign detection is one of the important tasks.Existing target detection networks are difficult to adapt to scenarios where target sizes are seriously imbalanced,and traffic sign targets are small and have unclear features,which makes detection more difficult.Therefore,we propose aHybrid Feature Fusion Traffic Sign detection algorithmbased onYOLOv7(HFFTYOLO).First,a self-attention mechanism is incorporated at the end of the backbone network to calculate feature interactions within scales;Secondly,the cross-scale fusion part of the neck introduces a bottom-up multi-path fusion method.Design reuse paths at the end of the neck,paying particular attention to cross-scale fusion of highlevel features.In addition,we found the appropriate channel width through a lot of experiments and reduced the superfluous parameters.In terms of training,a newregression lossCMPDIoUis proposed,which not only considers the problem of loss degradation when the aspect ratio is the same but the width and height are different,but also enables the penalty term to dynamically change at different scales.Finally,our proposed improved method shows excellent results on the TT100K dataset.Compared with the baseline model,without increasing the number of parameters and computational complexity,AP0.5 and AP increased by 2.2%and 2.7%,respectively,reaching 92.9%and 58.1%.

Artificial Immune Detection for Network Intrusion Data Based on Quantitative Matching Method

Artificial immune detection can be used to detect network intrusions in an adaptive approach and proper matching methods can improve the accuracy of immune detection methods.This paper proposes an artificial immune detection model for network intrusion data based on a quantitative matching method.The proposed model defines the detection process by using network data and decimal values to express features and artificial immune mechanisms are simulated to define immune elements.Then,to improve the accuracy of similarity calculation,a quantitative matching method is proposed.The model uses mathematical methods to train and evolve immune elements,increasing the diversity of immune recognition and allowing for the successful detection of unknown intrusions.The proposed model’s objective is to accurately identify known intrusions and expand the identification of unknown intrusions through signature detection and immune detection,overcoming the disadvantages of traditional methods.The experiment results show that the proposed model can detect intrusions effectively.It has a detection rate of more than 99.6%on average and a false alarm rate of 0.0264%.It outperforms existing immune intrusion detection methods in terms of comprehensive detection performance.

HWD-YOLO:A New Vision-Based Helmet Wearing Detection Method

It is crucial to ensure workers wear safety helmets when working at a workplace with a high risk of safety accidents,such as construction sites and mine tunnels.Although existing methods can achieve helmet detection in images,their accuracy and speed still need improvements since complex,cluttered,and large-scale scenes of real workplaces cause server occlusion,illumination change,scale variation,and perspective distortion.So,a new safety helmet-wearing detection method based on deep learning is proposed.Firstly,a new multi-scale contextual aggregation module is proposed to aggregate multi-scale feature information globally and highlight the details of concerned objects in the backbone part of the deep neural network.Secondly,a new detection block combining the dilate convolution and attention mechanism is proposed and introduced into the prediction part.This block can effectively extract deep featureswhile retaining information on fine-grained details,such as edges and small objects.Moreover,some newly emerged modules are incorporated into the proposed network to improve safety helmetwearing detection performance further.Extensive experiments on open dataset validate the proposed method.It reaches better performance on helmet-wearing detection and even outperforms the state-of-the-art method.To be more specific,the mAP increases by 3.4%,and the speed increases from17 to 33 fps in comparison with the baseline,You Only Look Once(YOLO)version 5X,and themean average precision increases by 1.0%and the speed increases by 7 fps in comparison with the YOLO version 7.The generalization ability and portability experiment results show that the proposed improvements could serve as a springboard for deep neural network design to improve object detection performance in complex scenarios.

SDH-FCOS:An Efficient Neural Network for Defect Detection in Urban Underground Pipelines

Urban underground pipelines are an important infrastructure in cities,and timely investigation of problems in underground pipelines can help ensure the normal operation of cities.Owing to the growing demand for defect detection in urban underground pipelines,this study developed an improved defect detection method for urban underground pipelines based on fully convolutional one-stage object detector(FCOS),called spatial pyramid pooling-fast(SPPF)feature fusion and dual detection heads based on FCOS(SDH-FCOS)model.This study improved the feature fusion component of the model network based on FCOS,introduced an SPPF network structure behind the last output feature layer of the backbone network,fused the local and global features,added a top-down path to accelerate the circulation of shallowinformation,and enriched the semantic information acquired by shallow features.The ability of the model to detect objects with multiple morphologies was strengthened by introducing dual detection heads.The experimental results using an open dataset of underground pipes show that the proposed SDH-FCOS model can recognize underground pipe defects more accurately;the average accuracy was improved by 2.7% compared with the original FCOS model,reducing the leakage rate to a large extent and achieving real-time detection.Also,our model achieved a good trade-off between accuracy and speed compared with other mainstream methods.This proved the effectiveness of the proposed model.