Poor Detection Performance Research Articles

SNR analysis of a multi-channel temporal correlation scheme in quantum-enhanced target detection.

In lossy and noisy environments, quantum-enhanced target detection based on temporal quantum correlation encounters low signal-to-noise ratio (SNR), resulting in poor detection performance. To address these challenges, we propose a multi-channel temporal correlation scheme. In this scheme, signal photons from multiple independent entangled sources illuminate the target and arrive at the same detector. Coincidences are obtained by correlation measurements of the entangled photons on one signal path and different reference paths. We then propose a weighted average processing method for fusing the coincidences to obtain higher SNR. The relationship between the SNR and the number of sources is analyzed for different background noise levels. It is shown that the SNR increases as the number of sources increases, but eventually approaches a limit. Experimental results verify the correction of our theoretical analysis.

Performance Analysis of Cooperative Spectrum Sensing using Empirical Mode Decomposition and Artificial Neural Network in Wireless Regional Area Network

Background: Radio spectrum is natural and the most precious means in wireless communication systems. Optimal spectrum utilization is a key concern for today's cutting-edge wireless communication networks. The impending problem of the lack of available spectrum has prompted the development of a new idea called “Cognitive Radio” (CR). Cooperative spectrum sensing (CSS) is utilized to improve the detection performance of the system. Several fusion algorithms of decision-making are proposed for sensing the licensed user, but they do not work well under low signal-to-noise ratio (SNR). Objectives: To address the issue of poor detection performance under low SNR, Empirical mode decomposition (EMD) and artificial neural network (ANN) based CSS under Rayleigh multipath fading channel in IEEE 802.22 wireless regional area network (WRAN) is proposed in this paper. Method: In this work, we propose the use of ANN as a fusion center. First, the received signal's energy is calculated using EMD. The computed energy, SNR, and false alarm probability are combined to form a data set of 2048 samples. They are utilized to train Levenberg- Marquardt back propagation training algorithm-based feed-forward neural network (FFNN). Using this trained network, CSS in WRAN is simulated under Rayleigh multipath fading. Results: Simulation results show that the proposed CSS method based on EMD-ANN outperforms the standard fast Fourier transform (FFT) and EMD detection-based cooperative spectrum sensing with a hard "OR" fusion at low SNR. With Pf =0.01, 100% detection accuracy with proposed techniques is obtained at SNR= -22dB. Conclusion: The findings show that the suggested approach outperforms EMD and FFT based energy detection scheme-based traditional CSS in low SNR environments.

EFS-YOLO: a lightweight network based on steel strip surface defect detection

Abstract With the advancement of deep learning technologies, industrial intelligent detection algorithms are gradually being applied in practical steel surface defect detection. Addressing the issues of high computational resource consumption and poor detection performance faced by existing models in large-scale industrial production lines, this paper proposes an EFS-YOLO (Efficient-Fast-Shared-YOLO) model based on improved YOLOv8s architecture. Firstly, the EfficientViT is employed as the feature extraction network, effectively reducing the model’s parameters and calculations while enhancing its capability to represent defect features. Secondly, the designed lightweight C2f-Faster-EffectiveSE Block (CFE-Block) was integrated into the model neck, accelerating feature fusion and better preserving detailed defect features at lower levels. Finally, the model detection head was reconstructed using the concept of shared parameters to address the high computational cost of the original detection head. Experimental results on the NEU-DET and GC10-DET datasets demonstrate that compared to the baseline model, the proposed model achieves a reduction in parameters, calculations and size by 49.5%, 62.7% and 46.9% respectively. It also exhibits an improvement in accuracy by 2.4% and 2.3% on the two datasets. The EFS-YOLO model effectively enhances precision in steel surface defect detection while maintaining lightweight characteristics, offering superior performance.

A novel robotic-assisted deep learning-enabled computer vision approach for nondestructive diagnosis of railway bolt faults

Abstract Railways play a vital role in the inland transportation system worldwide, and abnormal bolt components at the track joints are the main cause of train accidents. The detection and identification of faults in rail bolt components are of considerable research importance. To address this problem, numerous researchers have opted for computer vision-based methods to accomplish the detection and identification of the target, but the existing methods have poor detection performance diminished detection capabilities when the target position changes or some feature information is occluded, and the detection speed and accuracy are far from meeting the requirements of practical applications. Therefore, based on the construction of a dedicated dataset for bolt components, this paper uses the K-means dimensional clustering algorithm to re-cluster the dataset according to the target size characteristics, with the aim of reduce the bounding box regression error. At the same time, a novel loss function iteration method is proposed by incorporating an adaptive optimization algorithm, in order to improve the detection speed and ensure good convergence, and the model complexity is reduced based on deep model pruning. Finally, the optimized detection model is implemented on the robotic-assisted platform for testing, and the experimental results verify that the algorithm can quickly and accurately complete various fault diagnosis tasks of bolt components in practical applications. The main achievements of this study include the construction of a large-scale image dataset for novel rail bolt components and propelled the application of deep learning methods in vision-based rail bolt fault diagnosis.

MRD-YOLO: A Multispectral Object Detection Algorithm for Complex Road Scenes

Object detection is one of the core technologies for autonomous driving. Current road object detection mainly relies on visible light, which is prone to missed detections and false alarms in rainy, night-time, and foggy scenes. Multispectral object detection based on the fusion of RGB and infrared images can effectively address the challenges of complex and changing road scenes, improving the detection performance of current algorithms in complex scenarios. However, previous multispectral detection algorithms suffer from issues such as poor fusion of dual-mode information, poor detection performance for multi-scale objects, and inadequate utilization of semantic information. To address these challenges and enhance the detection performance in complex road scenes, this paper proposes a novel multispectral object detection algorithm called MRD-YOLO. In MRD-YOLO, we utilize interaction-based feature extraction to effectively fuse information and introduce the BIC-Fusion module with attention guidance to fuse different modal information. We also incorporate the SAConv module to improve the model’s detection performance for multi-scale objects and utilize the AIFI structure to enhance the utilization of semantic information. Finally, we conduct experiments on two major public datasets, FLIR_Aligned and M3FD. The experimental results demonstrate that compared to other algorithms, the proposed algorithm achieves superior detection performance in complex road scenes.

Real-Time Recognition Algorithm of Small Target for UAV Infrared Detection

Unmanned Aerial Vehicle (UAV) infrared detection has problems such as weak and small targets, complex backgrounds, and poor real-time detection performance. It is difficult for general target detection algorithms to achieve the requirements of a high detection rate, low missed detection rate, and high real-time performance. In order to solve these problems, this paper proposes an improved small target detection method based on Picodet. First, to address the problem of poor real-time performance, an improved lightweight LCNet network was introduced as the backbone network for feature extraction. Secondly, in order to solve the problems of high false detection rate and missed detection rate due to weak targets, the Squeeze-and-Excitation module was added and the feature pyramid structure was improved. Experimental results obtained on the HIT-UAV public dataset show that the improved detection model’s real-time frame rate increased by 31 fps and the average accuracy (MAP) increased by 7%, which proves the effectiveness of this method for UAV infrared small target detection.

SEB-YOLO: An Improved YOLOv5 Model for Remote Sensing Small Target Detection.

Due to the limited semantic information extraction with small objects and difficulty in distinguishing similar targets, it brings great challenges to target detection in remote sensing scenarios, which results in poor detection performance. This paper proposes an improved YOLOv5 remote sensing image target detection algorithm, SEB-YOLO (SPD-Conv + ECSPP + Bi-FPN + YOLOv5). Firstly, the space-to-depth (SPD) layer followed by a non-strided convolution (Conv) layer module (SPD-Conv) was used to reconstruct the backbone network, which retained the global features and reduced the feature loss. Meanwhile, the pooling module with the attention mechanism of the final layer of the backbone network was designed to help the network better identify and locate the target. Furthermore, a bidirectional feature pyramid network (Bi-FPN) with bilinear interpolation upsampling was added to improve bidirectional cross-scale connection and weighted feature fusion. Finally, the decoupled head is introduced to enhance the model convergence and solve the contradiction between the classification task and the regression task. Experimental results on NWPU VHR-10 and RSOD datasets show that the mAP of the proposed algorithm reaches 93.5% and 93.9%respectively, which is 4.0% and 5.3% higher than that of the original YOLOv5l algorithm. The proposed algorithm achieves better detection results for complex remote sensing images.

Non-Destructive Testing of a Fiber-Web-Reinforced Polymethacrylimide Foam Sandwich Panel with Terahertz Time-Domain Spectroscopy.

Terahertz (THz) non-destructive testing can detect internal defects in dielectric materials. However, this technology is mainly used for detecting thin and simple structures at present, lacking validations for the detection effectiveness of internal defects in thicker and more complex structures, such as fiber-web-reinforced composite sandwich panels. In this study, samples of fiber-web-reinforced polymethacrylimide foam sandwich panels, which are, respectively, 20 mm and 30 mm thick, were made to detect the internal debonding, inclusion, pore, and crack defects by the THz time-domain spectroscopy system (THz-TDS). The peak-to-peak-imaging algorithm, maximum-amplitude-imaging algorithm, minimum-amplitude-imaging algorithm, pulse-width-imaging algorithm, and time-of-flight-imaging algorithm were used to process and image the collected THz signals. The results showed that the peak-to-peak-imaging algorithm had the best performance. To address the low imaging resolution of THz-TDS, a block-based super-resolution reconstruction method-SSSRGAN-is proposed, which can improve image resolution while maintaining the clear edge contours of defects. The defect-detection results of the samples showed that THz-TDS could detect all pore, debonding, and crack defects, with a minimum size of 3 mm for pores and debonding and a minimum thickness of 1 mm for cracks. The method showed poor detection performance for inclusions with a thickness of 0.053 mm, but could still extract the defect features. Based on the THz-TDS reflection mode measurement principle, the thickness information of the panel, foam core, and web of the samples was calculated: the measurement error was no more than 0.870 mm for Sample #1 and no more than 0.270 mm for Sample #2, demonstrating the accuracy of THz-TDS in measuring the dimensions of sandwich panel structures. In general, THz technology shows potential for detecting internal defects and performing dimensional measurements in complex structures. With the advancement of portable devices and enhancements in detection speed, real-time on-site detection is anticipated in the future.

AiPE: A Novel Transformer-Based Pose Estimation Method

AiPE: A Novel Transformer-Based Pose Estimation Method

PE-Transformer: Path enhanced transformer for improving underwater object detection

Traditional object detection methods cannot effectively identify underwater objects with complex backgrounds, and it is difficult to fully obtain the details of small-scale underwater targets, resulting in poor detection performance. We propose a path-augmented Transformer detection framework to address these limitations to explore the semantic details of small-scale underwater targets in complex environments. On one hand, an embedded local path detection information scheme is devised to facilitate the interaction between high-level and low-level features, thereby enhancing the semantic representation of distinctive features of small-scale underwater targets. Rich dependency relationships are established between the acquired high-level and low-level features within the CSWin-Transformer framework, thus fortifying the semantic representation during the encoding phase. Furthermore, a individualized loss function is employed to optimize and fine-tune features at various hierarchical levels. On the other hand, a detection module with flexible and adaptive point representation that is different from conventional square detection methods is designed. This module covers the underwater target from any direction, and the salient point samples in classification localization and feature selection between points realize the feature selection improves the detection accuracy of underwater objects simultaneously. We designed a new weighted loss function to encourage the network to converge better. Experimental results on open-source underwater and remote sensing images of UTDAC, RUOD, and ADios show that the proposed method outperforms other underwater object detection methods in terms of precision(P), recall(R), comprehensive evaluation index of F1-score and FPS(frames per second).

Average effective subcarrier-domain sparse representation approach for target information estimation in CP-OFDM-based passive bistatic radar

Although some existing sparse representation (SR) methods are robust for target detection in passive bistatic radar (PBR), they still face the challenges of high computational complexity and poor detection performance for extremely low-signal-to-clutter ratio (SCR) target. So, an average effective subcarrier (AES)-domain sparse representation approach is investigated in this paper. Firstly, the AES-based SR model is proposed to solve the problem of high computational complexity, which is established by utilizing the sparseness of the orthogonal frequency-division multiplexing (OFDM) with cyclic prefix (CP) signals in each effective subcarrier domain. Then, considering the difficulty of detecting extremely low-SCR targets, clutter cancellation is implemented by the SR-based optimization model. Two AES-S algorithms, namely AES-S-based clutter cancellation in the time domain (AES-S-T) and AES-S-based clutter cancellation in the subcarrier domain (AES-S-C), are proposed, and the computational complexity is further reduced. Finally, extensive simulation and experimental results illustrate that the proposed algorithms have good detection performance and low computational complexity in PBR detection scene.

TFMFT: Transformer-based multiple fish tracking

Recent advancements in fish tracking methodologies provide valuable solutions for assessing fish growth, marine fisheries, and biological research. In particular, there has been a burgeoning interest in vision-based methods for fish tracking, owing to the enhanced computational capabilities facilitated by deep learning models. However, these methods face several challenges, including poor fish detection performance under complex backgrounds, the potential for identification switches caused by the non-rigid features and occlusions of fish, and the limited fault tolerance of extant approaches. In this paper, a transformer-based multiple fish tracking model (TFMFT) is proposed, specifically designed to address the issue of instance loss of fish targets in aquaculture ponds with complex background disturbance. In particular, we introduce a Multiple Association (MA) method that bolsters fault tolerance in tracking by synthesizing simple Intersection-over-Union matching in the identification (ID) matching module. Through empirical studies across diverse Transformer-based models, we comprehensively assessed the influence of architecture design on data requirements. Furthermore, to evaluate the performance and generalizability of fish tracking models, we present the Multiple_Fish_Tracking_2022 (MFT22) dataset. The results demonstrate that TFMFT achieves 30.6% IDF1 (Identification F-Score) on the MFT22 dataset, outperforming the state-of-the-art by 10.9% and showcasing superior performance over other models. The resources and pre-trained model will be available at: https://github.com/vranlee/TFMFT.

Enhanced sandwich immunoassay based on bivalent nanobody as an efficient immobilization approach for foodborne pathogens detection

BackgroundNanobodies (Nbs), which consist of only antigen-binding domains of heavy chain antibodies, have been used in a various range of applications due to their excellent properties. Nevertheless, the size of Nbs is so small that their antigen binding sites may be sterically hindered after random fixation as capture antibodies, thus leading to poor detection performance in immunoassays. To address this problem, we have focused on the multivalent modification of Nbs, wanted to retain the advantage of good stability through enlarging the size of Nbs to a certain extent, while improve its affinity and reduce its influence by spatial orientation. ResultsHere, we designed homo- and heterodimeric Nbs based on Nb413 and Nb422 which recognize different epitopes of Salmonella. The affinity of engineered bivalent nanobodies for S. Enteritidis were 2 orders of magnitude higher compared to monovalent Nbs and low to sub-nM KD, as calculated by Scatchard analysis. To further explore the potential of bivalent Nbs for the detection of Salmonella, we established a sandwich ELISA based on bivalent and phage-displayed Nbs (BNb-ELISA) for multiplex Salmonella determination. Compared with monovalent Nb-based ELISA, the limit of detection (LOD) of the BNb-ELISA was shown to increase 7.5-fold to 2.364 × 103 CFU mL−1 for S. Enteritidis. In addition, the feasibility of this approach for S. Enteritidis detection in real samples was evaluated, with recoveries ranging from 73.0 % to 125.6 % and coefficients of variation (CV) below 7.68 %. Significance and noveltyIn this study, we developed for the first time bivalent Nbs against Salmonella and examined their improved affinity and impact on the performance of ELISA assay. It confirmed the high binding affinity and good ability of dimeric Nbs to reduce the occupation of the binding sites of immobilized antibodies. Thus, the multivalent modification of Nbs was demonstrated to be a promising means to enhance the performance of Nbs-based immunoassays for foodborne pathogens.

BRITD: behavior rhythm insider threat detection with time awareness and user adaptation

Researchers usually detect insider threats by analyzing user behavior. The time information of user behavior is an important concern in internal threat detection. Existing works on insider threat detection fail to make full use of the time information, which leads to their poor detection performance. In this paper, we propose a novel behavioral feature extraction scheme: we implicitly encode absolute time information in the behavioral feature sequences and use a feature sequence construction method taking covariance into account to make our scheme adaptive to users. We select Stacked Bidirectional LSTM and Feedforward Neural Network to build a deep learning-based insider threat detection model: Behavior Rhythm Insider Threat Detection (BRITD). BRITD is universally applicable to various insider threat scenarios, and it has good insider threat detection performance: it achieves an AUC of 0.9730 and a precision of 0.8072 with the CMU CERT dataset, which exceeds all baselines.Graphical

Performance of the Action Observation-Based Brain-Computer Interface in Stroke Patients and Gaze Metrics Analysis.

Brain-computer interfaces (BCIs) are anticipated to improve the efficacy of rehabilitation for people with motor disabilities. However, applying BCI in clinical practice is still a challenge due to the great diversity of patients. In the current study, a novel action observation (AO) based BCI was proposed and tested on stroke patients. Ten non-hemineglect patients and ten hemineglect patients were recruited. Four AO stimuli were designed, each presenting a decomposed action to complete the reach-and-grasp task. EEG data and eye movement data were collected. Eye movement data was utilized to analyze the reasons for individual differences in BCI performance. Task discriminative component analysis was utilized to perform online target detection. The results showed that the designed AO-based BCI could simultaneously induce steady state motion visual evoked potential (SSMVEP) from the occipital region and sensory motor rhythm from the sensorimotor region in stroke patients. The average online detection accuracy among the four AO stimuli reached 67% within 3 s in the non-hemineglect group, while the accuracy only reached 35% in the hemineglect group. Gaze metrics showed that the average total duration of fixations during the stimulus phase in the hemineglect group was only 1.31 s ± 0.532 s which was significantly lower than that in the non-hemineglect group. The results indicated that hemineglect patients have difficulty gazing at the AO stimulus, resulting in poor detection performance and weak desynchronization in the sensorimotor region. Furthermore, the degree of neglect is inversely proportional to the target detection accuracy in hemineglect stroke patients. In addition, the gaze metrics associated with cognitive load were significantly correlated with the accuracy in non-hemineglect patients. It indicated the cognitive load may affect the AO-based BCI. The current study will expedite the clinical application of AO-based BCI.

MRFM: A timely detection method for DDoS attacks in IoT with multidimensional reconstruction and function mapping

To address the slow response time of existing detection modules to the Internet of Things (IoT) Distributed Denial of Service (DDoS) attacks, along with their low feature differentiation and poor detection performance, we propose MRFM, a timely detection method with multidimensional reconstruction and function mapping. Firstly, we employ a queue mechanism to capture and store incoming network traffic data within a predefined time frame. Subsequently, we introduce a multidimensional reconstruction neural network model, specifically designed to reconstruct quantitative features based on their respective indices by adjusting the loss function. This process is followed by the computation of multidimensional reconstruction errors and the transformation of vectors into mapping features, thereby augmenting the disparities among various types of traffic data and promoting the similarity within the same category of traffic data. Lastly, we extract frequency information from the qualitative feature matrix using information entropy calculations, enriching the feature profile of individual traffic instances. The experimental results on two benchmark datasets show that MRFM can effectively detect different types of DDoS attacks. Notably, MRFM consistently outperforms other existing methods, exhibiting an average metric improvement of up to 9.61 %.

Research on Forest Flame Detection Algorithm Based on a Lightweight Neural Network

To solve the problem of the poor performance of a flame detection algorithm in a complex forest background, such as poor detection performance, insensitivity to small targets, and excessive computational load, there is an urgent need for a lightweight, high-accuracy, real-time detection system. This paper introduces a lightweight object-detection algorithm called GS-YOLOv5s, which is based on the YOLOv5s baseline model and incorporates a multi-scale feature fusion knowledge distillation architecture. Firstly, the ghost shuffle convolution bottleneck is applied to obtain richer gradient information through branching. Secondly, the WIoU loss function is used to address the issues of GIoU related to model optimization, slow convergence, and inaccurate regression. Finally, a knowledge distillation algorithm based on feature fusion is employed to further improve its accuracy. Experimental results based on the dataset show that compared to the YOLOv5s baseline model, the proposed algorithm reduces the number of parameters and floating-point operations by approximately 26% and 36%, respectively. Moreover, it achieved a 3.1% improvement in mAP0.5 compared to YOLOv5s. The experiments demonstrate that GS-YOLOv5s, based on multi-scale feature fusion, not only enhances detection accuracy but also meets the requirements of lightweight and real-time detection in forest fire detection, commendably improving the practicality of flame-detection algorithms.

Aircraft sensor fault detection based on temporal two-dimensionalization

Aerial sensor fault detection is of great importance in flight missions. However, the dimensionality of sensor time-series data is extremely high and the time span is extremely long, which lead to poor detection performance of existing methods. To address these problems, this paper proposes a time-series to 2D fault detection (T2D) method for aerial sensor fault detection based on time-series. Firstly, the information entropy is applied to the classification and aggregation approximation algorithm to achieve effective compression of the data while fully retaining the time-series features. Secondly, the gramian angular field is introduced to encode the reduced-dimensional data into two-dimensional images, maintaining the long-range dependence of the original sequence. Finally, a flexible convolution block is designed and inserted into the encoder of the detection network Vision Transformer to improve the detection accuracy of the model. Experimental results show that the T2D model performs significantly better than other models on a simulated time-series dataset of a civilian aircraft test flight, indicating the effectiveness and superiority of the proposed method.

Characterization of acoustic detection efficiency using an unmanned surface vessel as a mobile receiver platform

Studies involving acoustic telemetry typically use stationary acoustic receivers arranged in an array or grid. Unmanned surface vehicle (USV)-based mobile receivers offer advantages over the latter approach: the USV can be programmed to autonomously carry a receiver to and from target locations, more readily adapting to a survey’s spatial scope and scale. This work examines the acoustic detection performance of a low-cost USV developed as a flexible sensing platform. The USV was fitted with an acoustic receiver and operated over multiple waypoints set at increasing distances from the transmitter in two modes: drifting and station-keeping. While drifting, the USV was allowed to drift from the waypoint; while station-keeping, the USV used its thruster to hold position. Detection performance of the USV was similar to that of stationary receivers while drifting, but significantly worse while station-keeping. Noise from the USV thruster was hypothesized as a potential cause of poor detection performance during station-keeping. Detection performance varied with the depth of the tethered receiver such that detection range was greater during the deepest (4.6 m) trials than during shallower (1.1 and 2.9 m) trials. These results provide insight and guidance on how a USV can be best used for acoustic telemetry, namely, navigating to a planned waypoint, drifting and lowering the receiver to a desired depth for listening, and then navigating to the next waypoint.

MCW: A Generalizable Deepfake Detection Method for Few-Shot Learning

With the development of deepfake technology, deepfake detection has received widespread attention. Although some deepfake forensics techniques have been proposed, they are still very difficult to implement in real-world scenarios. This is due to the differences in different deepfake technologies and the compression or editing of videos during the propagation process. Considering the issue of sample imbalance with few-shot scenarios in deepfake detection, we propose a multi-feature channel domain-weighted framework based on meta-learning (MCW). In order to obtain outstanding detection performance of a cross-database, the proposed framework improves a meta-learning network in two ways: it enhances the model’s feature extraction ability for detecting targets by combining the RGB domain and frequency domain information of the image and enhances the model’s generalization ability for detecting targets by assigning meta weights to channels on the feature map. The proposed MCW framework solves the problems of poor detection performance and insufficient data compression resistance of the algorithm for samples generated by unknown algorithms. The experiment was set in a zero-shot scenario and few-shot scenario, simulating the deepfake detection environment in real situations. We selected nine detection algorithms as comparative algorithms. The experimental results show that the MCW framework outperforms other algorithms in cross-algorithm detection and cross-dataset detection. The MCW framework demonstrates its ability to generalize and resist compression with low-quality training images and across different generation algorithm scenarios, and it has better fine-tuning potential in few-shot learning scenarios.