Detection Algorithm Research Articles

Ship detection in reefs and deep-sea with medium-high resolution images

ABSTRACT Accurate and efficient ship detection is crucial for ocean monitoring and management, especially in reefs and deep-sea, where fishing and illegal activities threaten sustainability of ecosystems. Obtaining the size of ships in reefs and deep-sea helps to identify ship types and to assess the impact of ships on marine ecosystems quantitatively. Ship detections were mainly applied at coast or river of the inland region using high spatial resolution remote sensing data due to its rich details and frequent coverage. However, the ships located in reefs and deep-sea regions were rarely studied because of price, and availability of high spatial resolution remote sensing data. Global covered medium-high resolution (10–30 m) remote sensing data (such as Sentinel-2) makes it possible to obtain the spatial-temporal distribution of ships, especially small ships. This study developed a deep learning ship detection algorithm – an enhanced Rotated-Ship Detector (RShipDet) to detect ships in reefs and deep-sea regions. RShipDet was applied to Nansha Islands based on two medium-high resolution ship detection datasets (Sentinel2-Ship and SDGSAT-Ship). These two datasets include various backgrounds of reefs and deep-sea, and complex scenarios such as cloud cover and parallel ships, adapting RShipDet to ship detection under complex circumstances in reefs and deep-sea. The results showed that: (1) Small gains: gains of 3.3% and 8.7% Average-Precision (AP) compared to Rotation-equivariant Detector (ReDet) and Faster-RCNN on Sentinel2-Ship dataset; (2) Strong generalization capabilities: 77.4% AP on SDGSAT-Ship dataset; (3) Better performance under complex conditions: RShipDet obtained more accurate ship detection results over regions with cloud cover, islands and reefs, deep-sea, and ports compared to other classical detectors. Our algorithm could be applied for better management of ocean resources and activities in reefs and deep-sea.

A lightweight Mini-YOLOv5s algorithm for small target crack detection in tunnels

ABSTRACT The automatic detection and identification of tunnel cracks is very important for safe driving and tunnel maintenance. In order to accelerate the application of automated detection and reduce the deployment cost of target detection algorithm on edge devices, this paper adopts non-destructive testing and proposes a lightweight Mini-YOLOv5s algorithm for small target crack detection in tunnels, which mainly includes DP-C3 module, GDConv module, ECA-C3 module, CARAFE lightweight upsampling operator and loss function IGIoU. Among them, the DP-C3 module uses PConv to reduce the computational complexity of the model, and uses dilated convolution to enrich the receptive field to improve the information loss. The GDConv module improves the feature extraction capability of deep information. The feature fusion of small crack target information is realised by using ECA-C3 module and CARAFE lightweight up-sampling operator. A new loss function IGIoU is used to pay more attention to the contact degree of the boundary frame, so as to minimise the loss and improve the detection accuracy. Experiments show that the detection accuracy of Mini-YOLOv5s on self-made tunnel crack dataset reaches 95%, which compensates for the information loss, while the number of model parameters and GFLOPs are reduced by 35.4% and 30.4% respectively.

An open dataset for individual tree detection in UAV LiDAR point clouds and RGB orthophotos in dense mixed forests

We present an open access dataset for development, evaluation, and comparison of algorithms for individual tree detection in dense mixed forests. The dataset consists of a detailed field inventory and overlapping UAV LiDAR and RGB orthophoto, which make it possible to develop algorithms that fuse multimodal data to improve detection results. Along with the dataset, we describe and implement a basic local maxima filtering baseline and an algorithm for automatically matching detection results to the ground truth trees for detection algorithm evaluation.

Construction of University Football Basic Teaching and Training System Based on Object Detection and Tracking Algorithm

Football, often referred to as soccer outside the UK and Europe, is a highly popular activity on many university campuses. As a result, football training programs are offered across many institutions, training students at a range of skill levels. Therefore, this paper investigates a deep learning (DL)-based algorithm for recognizing and tracking targets in sports training videos. The study specifically addresses those videos tailored to identifying small targets, by generating new multi-scale features and modifying anchor point generation rules. Experimental results demonstrate the algorithm's strong performance in tracking football targets. Compared to Histogram of Oriented Gradients, the DL-based model achieved a 29.58% increase in accuracy and a 39.68% decrease in error rates when recognizing football player movement features. This algorithm accurately locates the edge contours of a football player's movements, meaning that universities can, and should, actively reform football teaching and training to enhance teaching effectiveness by utilizing this powerful algorithm.

YOLO-TSF: A Small Traffic Sign Detection Algorithm for Foggy Road Scenes

YOLO-TSF: A Small Traffic Sign Detection Algorithm for Foggy Road Scenes

Identifying Light-curve Signals with a Deep-learning-based Object Detection Algorithm. II. A General Light-curve Classification Framework

Vast amounts of astronomical photometric data are generated from various projects, requiring significant effort to identify variable stars and other object classes. In light of this, a general, widely applicable classification framework would simplify the process of designing specific classifiers for various astronomical objects. We present a novel deep-learning framework for classifying light curves using a weakly supervised object detection model. Our framework identifies the optimal windows for both light curves and power spectra automatically, and zooms in on their corresponding data. This allows for automatic feature extraction from both time and frequency domains, enabling our model to handle data across different scales and sampling intervals. We train our model on data sets obtained from Kepler, TESS, and Zwicky Transient Facility multiband observations of variable stars and transients. We achieve an accuracy of 87% for combined variable and transient events, which is comparable to the performance of previous feature-based models. Our trained model can be utilized directly for other missions, such as the All-sky Automated Survey for Supernovae, without requiring any retraining or fine-tuning. To address known issues with miscalibrated predictive probabilities, we apply conformal prediction to generate robust predictive sets that guarantee true-label coverage with a given probability. Additionally, we incorporate various anomaly detection algorithms to empower our model with the ability to identify out-of-distribution objects. Our framework is implemented in the Deep-LC toolkit, which is an open-source Python package hosted on Github (https://github.com/ckm3/Deep-LC) and PyPI.

Ship target detection in SAR images based on SimAM attention YOLOv8

AbstractDeep learning has been widely applied in ship detection in synthetic aperture radar (SAR) imagery due to their powerful feature representation capabilities. However, YOLOv8 models treat all regions of the image equally during convolutional feature processing, resulting in less‐than‐ideal outcomes. To address this limitation, this study proposes a simple, parameter‐free attention module (SimAM) attention‐based YOLOv8 algorithm for ship detection in SAR images. The proposed algorithm first passes through a backbone network, which incorporates SimAM attention modules. The SimAM attention mechanism successfully allocates the convolutional neural network's 3D weights effectively using an energy function method, without introducing additional parameters. This mechanism enables the network to automatically emphasize key features in the image, enhancing its ability to represent target areas and suppress background interference. Subsequently, deep features are upsampled and fused with relatively shallow features to extract features at three different scales and achieve target detection, ultimately outputting classification and positional information of the targets. The effectiveness of the model on the SAR‐ship‐dataset is experimentally validated achieving an mAP50 value of 97.72% and an mAP50‐95 value of 68.99%, confirming the superiority of the proposed model.

Evaluating the Performance of Vascular Feature Extraction Boosting Algorithm for Peripheral Arterial Disease Detection

Evaluating the Performance of Vascular Feature Extraction Boosting Algorithm for Peripheral Arterial Disease Detection

A flexible perception method of thin smoke based on patch total bounded variation for buildings

Early fire warning is critical to the safety and stability of power systems. However, current methods encounter challenges in capturing subtle features, limiting their effectiveness in providing timely alerts for potential fire hazards. To overcome this drawback, a novel detection algorithm for thin smoke was proposed to enhance early fire detection capabilities. The core is that the Patch-TBV feature was proposed first, and the total bounded variation (TBV) was computed at the patch level. This approach is rooted in the understanding that traditional methods struggle to detect minute variations in image characteristics, particularly in scenarios where the features are dispersed or subtle. By computing TBV at a more localized level, the algorithm proposed gains a finer granularity in assessing image quality, enabling it to capture subtle variations that might indicate the presence of smoke or early signs of a fire. Another key aspect that sets our algorithm apart is the incorporation of subtle variation magnification. This technique serves to magnify subtle features within the image, leveraging the computed TBV values. This magnification strategy is pivotal for improving the algorithm’s precision in detecting subtle variations, especially in environments where smoke concentrations may be minimal or dispersed. To evaluate the algorithm’s performance in real-world scenarios, a comprehensive dataset, named TIP, comprising 3,120 images was constructed. The dataset covers diverse conditions and potential challenges that might be encountered in practical applications. Experimental results confirm the robustness and effectiveness of the proposed algorithm, showcasing its ability to provide accurate and timely fire warnings in various contexts. In conclusion, our research not only identifies the limitations of existing methods in capturing subtle features for early fire detection but also proposes a sophisticated algorithm, integrating Patch-TBV and micro-variation amplification, to address these challenges. The algorithm’s effectiveness and robustness are substantiated through extensive testing, demonstrating its potential as a valuable tool for enhancing fire safety in power systems and similar environments.

YOLOv8-E: An Improved YOLOv8 Algorithm for Eggplant Disease Detection

During the developmental stages, eggplants are susceptible to diseases, which can impact crop yields and farmers’ economic returns. Therefore, timely and effective detection of eggplant diseases is crucial. Deep learning-based object detection algorithms can automatically extract features from images of eggplants affected by diseases. However, eggplant disease images captured in complex farmland environments present challenges such as varying disease sizes, occlusion, overlap, and small target detection, making it difficult for existing deep-learning models to achieve satisfactory detection performance. To address this challenge, this study proposed an optimized eggplant disease detection algorithm, YOLOv8-E, based on You Only Look Once version 8 nano (YOLOv8n). Firstly, we integrate switchable atrous convolution (SAConv) into the C2f module to design the C2f_SAConv module, replacing some of the C2f modules in the backbone network of YOLOv8n, enabling our proposed algorithm to better extract eggplant disease features. Secondly, to facilitate the deployment of the detection model on mobile devices, we reconstruct the Neck network of YOLOv8n using the SlimNeck module, making the model lighter. Additionally, to tackle the issue of missing small targets, we embed the large separable kernel attention (LSKA) module within SlimNeck, enhancing the model’s attention to fine-grained information. Lastly, we combined intersection over union with auxiliary bounding box (Inner-IoU) and minimum point distance intersection over union (MPDIoU), introducing the Inner-MPDIoU loss to speed up convergence of the model and raise detection precision of overlapped and occluded targets. Ablation studies demonstrated that, compared to YOLOv8n, the mean average precision (mAP) and F1 score of YOLOv8-E reached 79.4% and 75.7%, respectively, which obtained a 5.5% increment and a 4.5% increase, while also reducing the model size and computational complexity. Furthermore, YOLOv8-E achieved higher detection performance than other mainstream algorithms. YOLOv8-E exhibits significant potential for practical application in eggplant disease detection.

A LiDAR-Camera Joint Calibration Algorithm Based on Deep Learning

Multisensor (MS) data fusion is important for improving the stability of vehicle environmental perception systems. MS joint calibration is a prerequisite for the fusion of multimodality sensors. Traditional calibration methods based on calibration boards require the manual extraction of many features and manual registration, resulting in a cumbersome calibration process and significant errors. A joint calibration algorithm for a Light Laser Detection and Ranging (LiDAR) and camera is proposed based on deep learning without the need for other special calibration objects. A network model constructed based on deep learning can automatically capture object features in the environment and complete the calibration by matching and calculating object features. A mathematical model was constructed for joint LiDAR-camera calibration, and the process of sensor joint calibration was analyzed in detail. By constructing a deep-learning-based network model to determine the parameters of the rotation matrix and translation matrix, the relative spatial positions of the two sensors were determined to complete the joint calibration. The network model consists of three parts: a feature extraction module, a feature-matching module, and a feature aggregation module. The feature extraction module extracts the image features of color and depth images, the feature-matching module calculates the correlation between the two, and the feature aggregation module determines the calibration matrix parameters. The proposed algorithm was validated and tested on the KITTI-odometry dataset and compared with other advanced algorithms. The experimental results show that the average translation error of the calibration algorithm is 0.26 cm, and the average rotation error is 0.02°. The calibration error is lower than those of other advanced algorithms.

Staining-Independent Malaria Parasite Detection and Life Stage Classification in Blood Smear Images

Malaria is a leading cause of morbidity and mortality in tropical and sub-tropical regions. This research proposed a malaria diagnosis system based on the you only look once algorithm for malaria parasite detection and the convolutional neural network algorithm for malaria parasite life stage classification. Two public datasets are utilized: MBB and MP-IDB. The MBB dataset includes human blood smears infected with Plasmodium vivax (P. vivax). While the MP-IDB dataset comprises 4 species of malaria parasites: P. vivax, P. ovale, P. malariae, and P. falciparum. Four distinct stages of life exist in every species, including ring, trophozoite, schizont, and gametocyte. For the MBB dataset, detection and classification accuracies of 0.92 and 0.93, respectively, were achieved. For the MP-IDB dataset, the proposed algorithms yielded the accuracies for detection and classification as follows: 0.84 and 0.94 for P. vivax; 0.82 and 0.93 for P. ovale; 0.79 and 0.93 for P. malariae; and 0.92 and 0.96 for P. falciparum. The detection results showed the models trained by P. vivax alone provide good detection capabilities also for other species of malaria parasites. The classification performance showed the proposed algorithms yielded good malaria parasite life stage classification performance. The future directions include collecting more data and exploring more sophisticated algorithms.

Inflation of crisis coverage? Tracking and explaining the changes in crisis labeling and crisis news wave salience 1785–2020

Abstract Has there been an inflation in crisis coverage in newspapers over the last centuries, and if so, what structural factors drive this change? We utilize computational text analyses along with our own signal detection algorithm to measure the presence of crisis keywords and the emergence of crisis news waves. An analysis of crisis coverage in The Times (U.K., 1785–2020, 183,239 news stories) shows that the share of coverage that uses crisis keywords has increased, though not steadily. The number and salience of crisis news waves tied to discernible events has increased at a slower pace. The hypothesized driving forces—government expansion, mediatization of politics, and the activity of crisis frame sponsors—explain the development well and allow accurate predictions even when ignoring time in the forecasting model. Crisis coverage seems to reflect not so much the problems society faces, but society’s identity, priorities, and outlook on the world.

Evaluation of droplet deposition parameters based on the Genetic-Otsu algorithm

Pesticide spraying is a cost-effective way to control crop pests and diseases. The effectiveness of this method relies on the deposition and distribution of the spray droplets within the targeted application area. There is a critical need for an accurate and stable detection algorithm to evaluate the liquid droplet deposition parameters on the water-sensitive paper (WSP) and reduce the impact of image noise. This study acquired 90 WSP samples with diverse coverage through field spraying experiments. The droplets on the WSP were subsequently isolated, and the coverage and density were computed, employing the fixed threshold method, the Otsu threshold method, and our Genetic-Otsu threshold method. Based on the benchmark of manually measured data, an error analysis was conducted on the accuracy of three methods, and a comprehensive evaluation was carried out. The relative error results indicate that the Genetic-Otsu method proposed in this research demonstrates superior performance in detecting droplet coverage and density. The relative errors of droplet density in the sparse, medium, and dense droplet groups are 2.7%, 1.5%, and 2.0%, respectively. The relative errors of droplet coverage are 1.5%, 0.88%, and 1.2%, respectively. These results demonstrate that the Genetic-Otsu algorithm outperforms the other two algorithms. The proposed algorithm effectively identifies small-sized droplets and accurately distinguishes the multiple independent contours of adjacent droplets even in dense droplet groups, demonstrating excellent performance. Overall, the Genetic-Otsu algorithm offered a reliable solution for detecting droplet deposition parameters on WSP, providing an efficient tool for evaluating droplet deposition parameters in UAV pesticide spraying applications.

SzCORE: Seizure Community Open-Source Research Evaluation framework for the validation of electroencephalography-based automated seizure detection algorithms.

The need for high-quality automated seizure detection algorithms based on electroencephalography (EEG) becomes ever more pressing with the increasing use of ambulatory and long-term EEG monitoring. Heterogeneity in validation methods of these algorithms influences the reported results and makes comprehensive evaluation and comparison challenging. This heterogeneity concerns in particular the choice of datasets, evaluation methodologies, and performance metrics. In this paper, we propose a unified framework designed to establish standardization in the validation of EEG-based seizure detection algorithms. Based on existing guidelines and recommendations, the framework introduces a set of recommendations and standards related to datasets, file formats, EEG data input content, seizure annotation input and output, cross-validation strategies, and performance metrics. We also propose the EEG 10-20 seizure detection benchmark, a machine-learning benchmark based on public datasets converted to a standardized format. This benchmark defines the machine-learning task as well as reporting metrics. We illustrate the use of the benchmark by evaluating a set of existing seizure detection algorithms. The SzCORE (Seizure Community Open-Source Research Evaluation) framework and benchmark are made publicly available along with an open-source software library to facilitate research use, while enabling rigorous evaluation of the clinical significance of the algorithms, fostering a collective effort to more optimally detect seizures to improve the lives of people with epilepsy.

Online transaction fraud detection in the banking sector using machine learning techniques

In this paper Nowadays, online transactions are increasing rapidly and payment scams are also increasing. A hacker is a person who gains access to another person's financial information. Improving feature selection is important to detect transaction risk in large, multidimensional data. Fraud detection and classification performance is divided into two parts: training data and test data. During this process we have focused on data analysis and pre-processing. In a transaction’s dataset, the predictive variables influence how well a machine learning algorithm for transaction risk identification classifies and identifies fraudulent transactions. This study examines the application of machine learning techniques in improving fraud detection in the banking industry. Using advanced algorithms and historical transaction data, the model aims to identify anomalous patterns that indicate fraudulent activity. Feature engineering, ensemble methods, and anomaly detection algorithms are used to improve the accuracy and performance of fraud detection systems. reducing false positives and provides an adaptive security mechanism. By harnessing the power of machine learning, this approach aims to strengthen the banking sectors defines against sophisticated fraud schemes. This study not only contributes to the academic understanding of fraud detection, but also provides industry practitioners with a practical framework for implementing proactive measures. In navigating the complex landscape of financial security, this research serves as a guide for building agile and responsive systems that stay ahead of the threats that evolve in the ever-changing landscape of the banking industry.

Entropic Regression Dynamic Mode Decomposition (ERDMD) Discovers Informative Sparse and Nonuniformly Time Delayed Models

In this work, we present a method that determines optimal multistep Dynamic Mode Decomposition (DMD) models via Entropic Regression (ER), which is a nonlinear information flow detection algorithm. Motivated by the Higher-Order DMD (HODMD) method of [Le Clainche & Vega, 2017], and the ER technique for network detection and model construction found in [Sun et al., 2015; AlMomani et al., 2020], we develop a method that we call ERDMD, which produces high fidelity time-delay DMD models that allow for nonuniformity in the delays. This optimal choice of delays is discovered by maximizing informativity as measured through ER. These models are shown to be highly efficient and robust. We test our method over several data sets generated by chaotic attractors and show that we are able to build excellent reconstructions using relatively minimal models. We likewise are able to better identify multiscale features via our models which enhances the utility of DMD.

Advancements in Electronic Component Assembly: Real-Time AI-Driven Inspection Techniques

This study presents an advanced methodology for improving electronic assembly quality through real-time, inline inspection utilizing state-of-the-art artificial intelligence (AI) and deep learning technologies. The primary goal is to ensure compliance with stringent manufacturing standards, notably IPC-A-610 and IPC-J-STD-001. Employing the existing infrastructure of pick-and-place machines, this system captures high-resolution images of electronic components during the assembly process. These images are analyzed instantly by AI algorithms capable of detecting a variety of defects, including damage, corrosion, counterfeit, and structural irregularities in components and their leads. This proactive approach shifts from conventional reactive quality assurance methods by integrating real-time defect detection and strict adherence to industry standards into the assembly process. With an accuracy rate exceeding 99.5% and processing speeds of about 5 ms per component, this system enables manufacturers to identify and address defects promptly, thereby significantly enhancing manufacturing quality and reliability. The implementation leverages big data analytics, analyzing over a billion components to refine detection algorithms and ensure robust performance. By pre-empting and resolving defects before they escalate, the methodology minimizes production disruptions and fosters a more efficient workflow, ultimately resulting in considerable cost reductions. This paper showcases multiple case studies of component defects, highlighting the diverse types of defects identified through AI and deep learning. These examples, combined with detailed performance metrics, provide insights into optimizing electronic component assembly processes, contributing to elevated production efficiency and quality.

Fault Prediction in Resistance Spot Welding: A Comparison of Machine Learning Approaches

Resistance spot welding is widely adopted in manufacturing and is characterized by high reliability and simple automation in the production line. The detection of defective welds is a difficult task that requires either destructive or expensive and slow non-destructive testing (e.g., ultrasound). The robots performing the welding automatically collect contextual and process-specific data. In this paper, we test whether these data can be used to predict defective welds. To do so, we use a dataset collected in a real industrial plant that describes welding-related data labeled with ultrasonic quality checks. We use these data to develop several pipelines based on shallow and deep learning machine learning algorithms and test the performance of these pipelines in predicting defective welds. Our results show that, despite the development of different pipelines and complex models, the machine-learning-based defect detection algorithms achieve limited performance. Using a qualitative analysis of model predictions, we show that correct predictions are often a consequence of inherent biases and intrinsic limitations in the data. We therefore conclude that the automatically collected data have limitations that hamper fault detection in a running production plant.

The Impact of land use change on peatland degradation: A case Nathong and Saming Village, Champhone district, Champasack Province, Lao PDR

The major impact on land use change and peatland degradation was directly from human activities and inappropriate management. Changing of land use and land cover (LULC) influence directly on peat land drained and degraded. Nathong and Saming Village, Champhone district, Champasack Province, Lao PDR is one of proposed peatland area and rapid LULC changes from 2016 due to a social-economic growth. Therefore, the assessment of land use change is required for sustainable peatland management. The main objectives of this study were to assess LULC and its change between 2011 to 2016 and 2016 to 2021. In this study, remote sensing data has been used for LULC classification using random forest classifier in Google Earth Engine (GEE). Then, the post-comparison change detection algorithm was applied for detected LULC changes. The result revealed that mixed deciduous area was continually decrees 523.44 ha and 607.05 ha from 2011 to 2016 and 2016 to 2021. Additionally, cassava and dry dipterocarp were increase from 138.24 ha in 2011 to 2016 to 386.1 ha in 2016 to 2021 and 93.15 ha in 2011 to 2016 to 156.42 ha in 2016 to 2021 respectively. Additionally, the derived overall accuracy and Kappa hat coefficients of the land use and land cover map in 2011,2016 and 2021 were higher than 85%. The average producer’s accuracy and user’s accuracy for all land use and land cover types were also more than 83%. The derived information of land use and land cover data of this study can be used as information to support decision-makers and land use planners for to awareness and sustainable peatland management.