Inaccurate Detection Research Articles

Boosting Point Set-Based Network with Optimal Transport Optimization for Oriented Object Detection

When handling complex remote sensing scenarios, rotational angle information can improve detection accuracy and enhance algorithm robustness, providing support for fine-grained detection. Point set representation is one of the most commonly used methods in arbitrary-oriented object detection tasks, leveraging discrete feature points to represent oriented targets and achieve high accuracy in angle prediction. However, due to the inherent discreteness of point set representation, it is prone to significant impact from isolated points and representational ambiguity in harsh application scenarios, leading to inaccurate detection. To address this issue, an efficient aerial object detector named BE-Det is proposed, which uses the optimal transport (OT) strategy to constrain the positions of isolated points. Additionally, a candidate point set quality evaluation scheme is designed to effectively assess the quality of candidate point sets. Experimental results on two challenging aerial datasets demonstrate that the proposed method outperforms several advanced detection methods.

Structural Plastic Damage Warning and Real-Time Sensing System Based on Cointegration Theory.

Structural damage can affect the long-term operation of equipment. Real-time damage warning for structures can effectively avoid accidents caused by structural damage. In this paper, a real-time warning method of structural plastic damage based on the cointegration theory is proposed. This method calculates the cointegration relationship between the strain signals at different measuring points, and the stability of the strain signal relationships is also evaluated. The problem of inaccurate detection caused by the error of strain measurement and environmental influence can be eliminated by the comprehensive judgment of strain between asymmetrical measuring points. A real-time damage sensing system is developed in this paper. In order to improve the real-time and practicability of the system, this paper proposes and determines the residual warning coefficient by analyzing the proportion of the strain residuals exceeding the residual threshold. The research on this sensing system has certain value for the engineering application of damage monitoring methods.

Target localization and defect detection of distribution insulators based on ECA‐SqueezeNet and CVAE‐GAN

AbstractInsulators, as typical equipment for distribution networks, provide good electrical insulation between live conductors and earth. Timely and accurate detection is essential for insulator detection issues. However, as the complexity of neural networks increases, the detection efficiency is often lower. Therefore, this paper proposes a fast insulator positioning and defect detection method. Firstly, for insulator target localization, the SqueezeNet network is improved using ECA attention mechanism. In addition, to address the issue of low defect detection accuracy, a joint algorithm has been proposed. The integration of convolutional variational autoencoder (CVAE) and generative adversarial network (GAN) solve their own shortcomings due to different image focus angles. The target localization accuracy reaches 94.30%, and the defect detection accuracy reaches 89.60%. It solves the problems of difficulty in locating small targets in a large field of view and inaccurate detection due to a small number of abnormal samples. This method has been tried and tested in practical distribution network systems.

Enhancing P-wave localization for accurate detection of second-degree and third-degree atrioventricular conduction blocks

Objective.This paper tackles the challenge of accurately detecting second-degree and third-degree atrioventricular block (AVB) in electrocardiogram (ECG) signals through automated algorithms. The inaccurate detection of P-waves poses a difficulty in this process. To address this limitation, we propose a reliable method that significantly improves the performances of AVB detection by precisely localizing P-waves.Approach.Our proposed P-WaveNet utilized an attention mechanism to extract spatial and temporal features, and employs a bidirectional long short-term memory module to capture inter-temporal dependencies within the ECG signal. To overcome the scarcity of data for second-degree and third-degree AVB (2AVB,3AVB), a mathematical approach was employed to synthesize pseudo-data. By combining P-wave positions identified by the P-WaveNet with key medical features such as RR interval rhythm and PR intervals, we established a classification rule enabling automatic AVB detection.Main results. The P-WaveNet achieved an F1 score of 93.62% and 91.42% for P-wave localization on the QT Dataset and Lobachevsky University dataset datasets, respectively. In the BUTPDB dataset, the F1 scores for P-wave localization in ECG signals with 2AVB and 3AVB were 98.29% and 62.65%, respectively. Across two independent datasets, the AVB detection algorithm achieved F1 scores of 83.33% and 84.15% for 2AVB and 3AVB, respectively.Significance.Our proposed P-WaveNet demonstrates accurate identification of P-waves in complex ECGs, significantly enhancing AVB detection efficacy. This paper's contributions stem from the fusion of medical expertise with data augmentation techniques and ECG classification. The proposed P-WaveNet demonstrates potential clinical applicability.

An Improved YOLOv7 Model for Surface Damage Detection on Wind Turbine Blades Based on Low-Quality UAV Images

The efficient damage detection of the wind turbine blade (WTB), the core part of the wind power, is very improtant to wind power. In this paper, an improved YOLOv7 model is designed to enhance the performance of surface damage detection on WTBs based on the low-quality unmanned aerial vehicle (UAV) images. (1) An efficient channel attention (ECA) module is imbeded, which makes the network more sensitive to damage to decrease the false detection and missing detection caused by the low-quality image. (2) A DownSampling module is introduced to retain key feature information to enhance the detection speed and accuracy which are restricted by low-quality images with large amounts of redundant information. (3) The Multiple attributes Intersection over Union (MIoU) is applied to improve the inaccurate detection location and detection size of the damage region. (4) The dynamic group convolution shuffle transformer (DGST) is developed to improve the ability to comprehensively capture the contours, textures and potential damage information. Compared with YOLOv7, YOLOv8l, YOLOv9e and YOLOv10x, this experiment’s results show that the improved YOLOv7 has the optimal detection performance synthetically considering the detection accuracy, the detection speed and the robustness.

A multi-channel spatial information feature based human pose estimation algorithm

Human pose estimation is an important task in computer vision, which can provide key point detection of human body and obtain bone information. At present, human pose estimation is mainly utilized for detection of large targets, and there is no solution for detection of small targets. This paper proposes a multi-channel spatial information feature based human pose (MCSF-Pose) estimation algorithm to address the issue of medium and small targets inaccurate detection of human key points in scenarios involving occlusion and multiple poses. The MCSF-Pose network is a bottom-up regression network. Firstly, an UP-Focus module is designed to expand the feature information while reducing parameter computation during the up-sampling process. Then, the channel segmentation strategy is adopted to cut the features, and the feature information of multiple dimensions is retained through different convolutional groups, which reduces the parameter lightweight network model and makes up for the loss of the feature information associated with the depth of the network. Finally, the three-layer PANet structure is designed to reduce the complexity of the model. With the aid of the structure, it also to improve the detection accuracy and anti-interference ability of human key points. The experimental results indicate that the proposed algorithm outperforms YOLO-Pose and other human pose estimation algorithms on COCO2017 and MPII human pose datasets.

Impact and Correction of an Anterior Phakic Intraocular Lens on Swept-Source Optical Coherence Tomography Biometry.

To evaluate the impact of anterior chamber phakic intraocular lens (pIOL) on swept-source optical coherence tomography (SS-OCT) biometric measurements and IOL power calculation. This retrospective analysis of 67 eyes of 49 patients with previous anterior chamber pIOL implantation analyzed the accuracy of automatic segmentation of the anterior surface of the crystalline lens and its impact on anterior chamber depth (ACD, measured from epithelium to lens), lens thickness measurements, and IOL power calculation. The sample was divided into two groups: correct detection of the anterior surface of the crystalline lens and inaccurate detection. Segmentation of eyes from the inaccurate detection group was manually corrected and ACD and lens thickness were calculated using ImageJ software. IOL power was calculated using 7 formulas for both measurements. The anterior surface of the crystalline lens was mis-identified in 13 (19.4%) eyes. ACD was underestimated (Δ -0.85 ± 0.33 mm, P < .001) and lens thickness was overestimated (Δ +0.81 ± 0.25 mm, P < .001). Manual correction changed the target spherical equivalent only in the Haigis formula (P = .009). After correction for segmentation bias, the Pearl DGS, Cooke K6, and EVO 2.0 formulas showed the lowest prediction error, with the Pearl DGS showing greatest accuracy within ±1.00 diopters of prediction error range (81.0%). SS-OCT biometry misidentifies the anterior surface of the crystalline lens in a significant proportion, resulting in significant IOL power calculation error in the Haigis formula. Manual proofing of segmentation is mandatory in every patient with anterior chamber pIOL implantation. After correct segmentation, the Pearl DGS, Cooke K6, and EVO seem to be the best formulas. [J Refract Surg. 2024;40(8):e562-e568.].

Damage detection of frame structure using a novel time-domain regression method

Shear structure model is the most frequently used to model for the damage detection of frame building structures. However, due to the existence of modelling error, using a shear structure model to perform damage detection of a complex frame structure often results in inaccurate detection results. In this paper, a novel reduced model for the frame is proposed, which converts a multi-story multi-bay plane frame into a beam-like model, having one translational and two rotational degrees-of-freedom for each floor. Based on the new model, a novel time-domain regression method (TDRM) was established using the spectral density function between the horizontal acceleration of the frame floor and the reference response to identify the equivalent layer stiffness and damping parameters. Finally, a five-story two-bay frame structure is used to demonstrate the efficacy of the proposed time-domain regression method of estimating structural parameters and identifying structural damage.The results show that this method can identify, locate, and quantify the structural stiffness changes accurately.

Intelligent Decision Support System in Healthcare using Machine Learning Models

Background: The use of intelligent decision support systems (IDSS) is widespread in the healthcare industry, particularly for real-time data, client and family history datasets, and prevalent patient features. Objective: A massive chunk of various kinds of health data sets, including sensor information, medical evidence, and omic statistics, are produced by the modern techniques in this field and eventually transferred to a machine learning (ML) element for extracting data, categorization, as well as mining. Method: In recent times, many patents have been focused on healthcare monitoring; however, they do not adequately incorporate appropriate algorithms for data collection, analysis, and prediction. The data collected is used for predictive modelling, then additionally, machine learning techniques are assisting to compare acquired datasets mathematically for decision-making platforms that may learn to recognise the recent trend and anticipated future problems. Depending on the dataset type, ML-based techniques can assess the circumstances. Training datasets are crucial for correctly anticipating both current and emerging events as well as new challenges. Results: Since the importance of data acquisition determines how well learning models function, any deformed data of the types of dirty data, noisy data, unstructured data, and inadequate information results in inaccurate detection, estimate, and prediction. Conclusion: Additionally, in contrast to other approaches, the experimental findings demonstrate the usefulness of the proposed method as a widespread implementation of machine learning algorithms within healthcare systems.

A dynamic thermal sensing mechanism with reconfigurable expanded-plane structures

The precise measurement of temperature is crucial in various fields such as biology, medicine, industrial automation, energy management, and daily life applications. While in most scenarios, sensors with a fixed thermal conductivity inevitably mismatch the analogous parameter of the medium being measured, thus causing the distortion and inaccurate detection of original temperature fields. Despite recent efforts on addressing the parameter-mismatch issue, all current solutions are constrained to a fixed working medium, whereas a more universal sensor should function in a variety of scenes. Here, we report a dynamic and reconfigurable thermal sensor capable of highly accurate measurements in diverse working environments. Remarkably, thanks to the highly tunable thermal conductivity of the expanded-plane structure, this sensor works effectively on background mediums with a wide range of conductivity. Such a development greatly enhances the robustness and adaptability of thermal sensors, setting a solid foundation for applications in multi-physical sensing scenarios.

Machine vision-based autonomous road hazard avoidance system for self-driving vehicles

The resolution of traffic congestion and personal safety issues holds paramount importance for human’s life. The ability of an autonomous driving system to navigate complex road conditions is crucial. Deep learning has greatly facilitated machine vision perception in autonomous driving. Aiming at the problem of small target detection in traditional YOLOv5s, this paper proposes an optimized target detection algorithm. The C3 module on the algorithm’s backbone is upgraded to the CBAMC3 module, introducing a novel GELU activation function and EfficiCIoU loss function, which accelerate convergence on position loss lbox, confidence loss lobj, and classification loss lcls, enhance image learning capabilities and address the issue of inaccurate detection of small targets by improving the algorithm. Testing with a vehicle-mounted camera on a predefined route effectively identifies road vehicles and analyzes depth position information. The avoidance model, combined with Pure Pursuit and MPC control algorithms, exhibits more stable variations in vehicle speed, front-wheel steering angle, lateral acceleration, etc., compared to the non-optimized version. The robustness of the driving system's visual avoidance functionality is enhanced, further ameliorating congestion issues and ensuring personal safety.

Deep Learning-Based Acoustic Emission Signal Filtration Model in Reinforced Concrete

AbstractAcoustic emission is a nondestructive testing (NDT) technique, widely used to monitor the condition of structures for safety reasons especially in real time. The method utilizes the electrical signals generated by the elastic waves in a material under load to detect and locate damage in structures. However, identifying the sources of AE signals in concrete or composite materials can be challenging due to the anisotropic properties of materials and interpreting a large amount of AE data, leading to data misinterpretation and inaccurate detection of damage. Hence, the need for filtering out noise-induced signals from recorded data and emphasizing the actual AE source is crucial for monitoring and source localization of damage in real time. This study proposed a one-dimensional convolutional neural network (1D-CNN) deep learning approach to filter around 22,000 AE data in a reinforced concrete (RC) beam. The model utilizes significant AE parameters identified through neighborhood component analysis (NCA) to classify true AE signals from noise-induced signals. By using the optimized network parameters, a high classification accuracy of 97% and 96.29% was achieved during the training and testing phases, respectively. To check the reliability of the proposed AE filtering model in the real world, it was evaluated and verified using source location AE activities collected during a four-point bending test on a shear-deficient beam. The outcomes suggest that the proposed AE filtration model has the potential to accurately classify AE signals with an accuracy of 92.8% and proved that the filtration model provides accurate and valuable insight into source location determination.

Phage-based magnetic capture method as an aid for real-time recombinase polymerase amplification detection of Salmonella spp. in milk

Salmonella is a major cause of foodborne diseases worldwide. Conventional rapid assays for detecting Salmonella in real samples often encounter severe matrix interference or detect a limited number of species of a genus, resulting in inaccurate detection. In this study, we developed a method that combined phage-based magnetic capture with real-time recombinase polymerase amplification (RPA) for the rapid, highly sensitive, and specific detection of Salmonella in milk with an ultra-low detection limit. The Felix O-1 phage-conjugated magnetic beads (O-1 pMBs) synthesized in this method showed excellent capture ability for Salmonella spp. and ideal specificity for non-Salmonella strains. After O-1 pMBs-based magnetic separation, the limit of detection of the real-time RPA assay was 50 cfu/mL in milk samples, which was significantly increased by a magnitude of 3 to 4 orders. The method exhibited a high sensitivity (compatibility) of 100% (14/14) for all tested Salmonella serotype strains and an ideal specificity (exclusivity) of 100% (7/7) for the tested non-Salmonella strains. The entire detection process, including Salmonella capture, DNA extraction, and real-time RPA detection, was completed within 1.5 h. Furthermore, milk samples spiked with 10 cfu/25 mL of Salmonella were detected positive after being cultured in buffered peptone water for only 3 h. Therefore, the proposed method could be an alternative for the rapid and accurate detection of Salmonella.

An effective segmentation and attention-based reptile residual capsule auto encoder for pest classification.

Insect pests are a major global factor affecting agricultural crop productivity and quality. Rapid and precise insect pest detection is crucial for improving handling and prediction techniques. There are several methods for pest detection and classification tasks; still, the inaccurate detection, computation complexity and several other challenges affect the performance of the model. Thus, this research presents a Deep Learning (DL) approach that has led to significant advancements and is currently being applied successfully in many domains, such as autonomous insect pest detection. Initially, the input images are gathered from the test dataset. The next step in pre-processing the input images is to improve the model capacity by removing unwanted data using the Enhanced Kuan filter method. Then, the pre-processed images are segmented using the Attention-based U-Net method. Finally, a novel Attention Based Reptile Residual Capsule Auto Encoder (ARRCAE) technique is proposed to classify and recognize crop pests. Furthermore, the Improved Reptile Search Optimisation (IRSO) algorithm is employed to fine-tune the classification parameters optimally. As a result, the proposed study enhances performance by classifying crop pest detection systems. The suggested method makes use of a Python tool for simulation, and pest datasets are utilized for result analysis. The suggested model beats other current models with an accuracy of 98%, precision of 97%, recall of 96%, and specificity of 99% for the pest dataset, per the simulation results that were obtained. © 2024 Society of Chemical Industry.

Research on improved small scale face detection algorithm based on Yolov5

To address the challenges of dense scenarios with densely distributed small-scale faces, severe occlusions, and unclear features leading to inaccurate detection and high miss rates, we propose a lightweight small-scale face detection algorithm based on YOLOv5. The aim is to enhance the accuracy and precision of target detection. Firstly, we introduce the Convolutional Block Attention Module (CBAM) into the existing backbone network, obtaining more detailed features by comprehensively considering both spatial and channel dimensions. Next, in the Neck network, we embed involution to enhance channel information and weight distribution. Finally, a new feature fusion layer is added to improve the capture capability of feature information for smaller pixels and smaller targets in visible areas by integrating deep semantic information with shallow semantic information. The experimental results demonstrate that the improved model exhibits an increase in the average precision across all three subsets of the public WIDER FACE dataset, with improvements of 3.2%, 3.4%, and 2.6% respectively. The detection frame rate reaches 87 frames per second (FPS), significantly enhancing the detection performance of facial targets. This improvement meets the accuracy and real-time requirements for detecting small-scale facial targets in dense scenarios.

DBH-YOLO: a surgical instrument detection method based on feature separation in laparoscopic surgery.

Accurately locating and analysing surgical instruments in laparoscopic surgical videos can assist doctors in postoperative quality assessment. This can provide patients with more scientific and rational solutions for healing surgical complications. Therefore, we propose an end-to-end algorithm for the detection of surgical instruments. Dual-Branched Head (DBH) and Overall Intersection over Union Loss (OIoU Loss) are introduced to solve the problem of inaccurate surgical instrument detection, both in terms of localization and classification. An effective method (DBHYOLO) for the detection for laparoscopic surgery in complex scenarios is proposed. This study manually annotates a new laparoscopic gastric cancer resection surgical instrument location dataset LGIL, which provides a better validation platform for surgical instrument detection methods. The proposed method's performance was tested using the m2cai16-tool-locations, LGIL, and Onyeogulu datasets. The mean Average Precision (mAP) values obtained were 96.8%, 95.6%, and 98.4%, respectively, which were higher than the other classical models compared. The improved model is more effective than the benchmark network in distinguishing between surgical instrument classes with high similarity and avoiding too many missed detection cases. In this paper, the problem of inaccurate detection of surgical instruments is addressed from two different perspectives: classification and localization. And the experimental results on three representative datasets verify the performance of DBH-YOLO. It is shown that this method has a good generalization capability.

Multi-Granularity Aggregation with Spatiotemporal Consistency for Video-Based Person Re-Identification.

Video-based person re-identification (ReID) aims to exploit relevant features from spatial and temporal knowledge. Widely used methods include the part- and attention-based approaches for suppressing irrelevant spatial-temporal features. However, it is still challenging to overcome inconsistencies across video frames due to occlusion and imperfect detection. These mismatches make temporal processing ineffective and create an imbalance of crucial spatial information. To address these problems, we propose the Spatiotemporal Multi-Granularity Aggregation (ST-MGA) method, which is specifically designed to accumulate relevant features with spatiotemporally consistent cues. The proposed framework consists of three main stages: extraction, which extracts spatiotemporally consistent partial information; augmentation, which augments the partial information with different granularity levels; and aggregation, which effectively aggregates the augmented spatiotemporal information. We first introduce the consistent part-attention (CPA) module, which extracts spatiotemporally consistent and well-aligned attentive parts. Sub-parts derived from CPA provide temporally consistent semantic information, solving misalignment problems in videos due to occlusion or inaccurate detection, and maximize the efficiency of aggregation through uniform partial information. To enhance the diversity of spatial and temporal cues, we introduce the Multi-Attention Part Augmentation (MA-PA) block, which incorporates fine parts at various granular levels, and the Long-/Short-term Temporal Augmentation (LS-TA) block, designed to capture both long- and short-term temporal relations. Using densely separated part cues, ST-MGA fully exploits and aggregates the spatiotemporal multi-granular patterns by comparing relations between parts and scales. In the experiments, the proposed ST-MGA renders state-of-the-art performance on several video-based ReID benchmarks (i.e., MARS, DukeMTMC-VideoReID, and LS-VID).

Attention guided multi-level feature aggregation network for camouflaged object detection

Camouflaged object detection (COD) aims to identify objects that are visually blended into their highly similar surroundings, which is an extremely complex and challenging visual task in real-world scenarios, and has recently attracted increasing research interest in the field of computer vision due to its valuable applications. The existing deep learning based methods of COD have the following problems: 1) the ambiguous boundary of the camouflaged objects in prediction map, 2) an inaccurate detection of the camouflaged object with accurate and complete structure details. To this end, an attention guided multi-level feature aggregation network is proposed for this task, which is based on three key designs. First, by embedding spatial pyramid attention (SPA) in ResNet-like backbone network, better multi-level features are extracted to identify the camouflaged objects with complete internal details. Second, an edge context module is designed to make full use of edge information, which highlights camouflaged object structure and generates accurate edge localization of COD. Third, the feature aggregation module based on local attention is used to fuse these enhanced multi-level features and edge context cues, which consists of two major components: the iterative Attentional Feature Fusion (iAFF) module and the Dual-branch Global Context Module (DGCM). Compared with the existing 16 state-of-the-art methods, extensive experiments on four widely-used benchmark datasets under four authoritative evaluation metrics illustrate that the proposed method is very beneficial to the COD task.

MGSGA: Multi-grained and Semantic-Guided Alignment for Text-Video Retrieval

In the text-video retrieval task, the objective is to calculate the similarity between a text and a video, and rank the relevant candidates higher. Most existing methods only consider the text-video semantic alignment in the global view. But using mean-pooling to obtain global semantics and simply aligning text and video in the global view may lead to semantic bias. In addition, some methods utilize offline object detectors or sentence parsers to obtain entity-level information in text and video and achieve local alignment. However, inaccurate detection introduces possible errors and such approaches prevent models from being trained end-to-end for retrieval. To overcome these limitations, we propose multi-grained and semantic-guided alignment for text-video retrieval in this paper, which can achieve fine-grained alignment based on video frames and text words, local alignment based on semantic centers, and global alignment. Specially, we explore summary semantics of text and video to guide the local alignment based on semantic centers for we believe that the importance of each semantic center is determined by summary semantics. We evaluate our approach on four benchmark datasets of MSRVTT, MSVD, ActivityNet Captions, and DiDeMo, achieving better performance than most existing methods.

Instrument-free and rapid Vibrio parahaemolyticus biosensing based on coupled dual-signal of paper-needle and LnMOF

At present, most reported studies on paper-based sensing devices for food-borne pathogens usually rely on a single signal as the basis for judging the results, accompanied with inaccurate detection results. In this study, we used the lanthanide MOFs being covalently modified with aptamers (Apt-LnMOFs) as the sensing and recognition element of Vibrio parahaemolyticus (VP), and the paper-based needle shape device (paper-needle) as the visual signal output carrier. Finally, an accurate detection strategy for VP was successfully constructed. In terms of signal output, the two signal styles of Apt-LnMOFs diffusion distance on the paper-needle and R value (measured by the RGB analysis software of smartphone) are coupled to jointly improve the accuracy of detection results. The method is simple and fast, and the detection accuracy in real samples is 91%. This strategy is very suitable for on-site rapid detection of pathogens, and does not require large professional instruments.