Automatic Detection Method Research Articles

An efficient CNN-based detector for photovoltaic module cells defect detection in electroluminescence images

Electroluminescence (EL) imaging provides a high spatial resolution for inspecting photovoltaic (PV) cells, enabling the detection of various types of PV cell defects. Recently, convolutional neural network (CNN) based automatic detection methods for PV cell defects using EL images have attracted much attention. However, existing methods struggle to achieve a good balance between detection accuracy and efficiency. To address this issue, we propose a novel method for efficient PV cell defect detection. Firstly, we utilize Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm to improve EL image contrast, making defect features become more distinguishable. Secondly, we propose a lightweight defect detector using EfficientNet-B0 as its backbone. Moreover, we design a graph channel attention module (GCAM) to improve CNN’s limitation in modeling global information. It executes graph channel reasoning to generate enriched feature representation beyond the local receptive field, which is beneficial for distinguishing PV cell defects with similar local details. Next, we utilize focal loss to train the detector, enhancing its ability to detect challenging defects. Lastly, the proposed method is evaluated on the PVEL dataset and it achieved an accuracy of 97.81%, precision of 97.70%, recall of 97.59%, F1-score of 97.64%, and MCC of 97.32%, demonstrating our method is effective and outperforms state-of-the-art methods across various metrics.

The evaluation of hand-crafted and learned-based features in Terrestrial Laser Scanning-Structure-from-Motion (TLS-SfM) indoor point cloud registration: the case study of cultural heritage objects and public interiors

Modern technologies are commonly used to inventory different architectural or industrial objects (especially cultural heritage objects and sites) to generate architectural documentation or 3D models. The Terrestrial Laser Scanning (TLS) method is one of the standard technologies researchers investigate for accurate data acquisition and processing required for architectural documentation. The processing of TLS data to generate high-resolution architectural documentation is a multi-stage process that begins with point cloud registration. In this step, it is a common practice to identify corresponding points manually, semi-manually or automatically. There are several challenges for the TLS point cloud processing in the data registration process: correct spatial distribution, marking of control points, automation, and robustness analysis. This is particularly important when large, complex heritage sites are investigated, where it is impossible to distribute marked control points. On the other hand, when orientating multi-temporal data, there is also the problem of corresponding reference points. For this reason, it is necessary to use automatic tie-point detection methods. Therefore, this article aims to evaluate the quality and completeness of the TLS registration process using 2D raster data in the form of spherical images and Affine Hand-crafted and Learned-based detectors in the multi-stage TLS point cloud registration as test data; point clouds were used for the historic 17th-century cellars of the Royal Castle in Warsaw without decorative structures, two baroque rooms in the King John III Palace Museum in Wilanów with decorative elements, ornaments and materials on the walls and flat frescoes, and two modern test fields, narrow office, and empty shopping mall. The extended Structure-from-Motion was used to determine the tie points for the complete TLS registration and reliability analysis. The evaluation of detectors demonstrates that for the test sites exhibiting rich textures and numerous ornaments, a combination of AFAST, ASURF, ASIFT, SuperGlue and LoFTR can be effectively employed. For the point cloud registration of less textured buildings, it is advisable to use AFAST/ASIFT. The robust method for point cloud registration exhibits comparable outcomes to the conventional target-based and Iterative Closest Points methods.

The climatology and nature of warm-season convective cells in cold-frontal environments over Germany

Abstract. Cold fronts provide an environment particularly favourable for convective initiation in the mid-latitudes and can also be associated with convective hazards such as flooding, wind, hail and lightning. We build a climatology of cold-frontal convective cells between 2007–2016 for April–September in a cell-front distance framework by combining a radar-based cell detection and tracking dataset and automatic front detection methods applied to reanalysis data. We find that on average around twice as many cells develop on cold-frontal cell days compared to non-cold-frontal cell days. Using the 700 hPa level as a reference point, we show the maximum cell frequency is 350–400 km ahead of the 700 hPa front, which is marginally ahead of the typical surface front location. The 700 hPa front location marks the minimum cell frequency and a clear shift in regime between cells, with a weakened diurnal cycle on the warm side of the 700 hPa cold front and strongly diurnally driven cells on the cold side of the 700 hPa front. High cell frequency is found several hundreds of kilometres ahead of the surface front, and cells in this region are most likely to be associated with mesocyclones, intense convective cores and lightning. Namely, mesocyclones were detected in around 5.0 % of pre-surface-frontal cells compared to only 1.5 % of non-cold-frontal cells. The findings in this study are an important step towards a better understanding of cold-frontal convection climatology and links between cold fronts and convective hazards.

Automatic histogram-based defect detection in glass fibre reinforced polymer composites using terahertz time-domain spectroscopy reflection imaging

We developed a method for automatic detection of defects in glass fibre reinforced polymer (GFRP) based on a terahertz time-domain spectroscopy (TDS) system. Two GFRP specimens with round defects of different sizes (inclusions and delaminations) were measured by means of a high-speed TDS reflective scanner with an electronically controlled optical sampling scheme. The algorithm provided a series of C-scans, which were then pre-selected to identify only those with features potentially associated with defects. A thresholding method based on Gaussian fitting of the histogram was developed. The results showed that 87 % of defects were detected for the specimen with inclusions and 100 % for the specimen with delaminations. The algorithm also provided information about the depth of defects, as each C-scan was assigned to a time slice. The results demonstrate that the TDS-based system, along with the proposed histogram-based method, can automatically detect defects in GFRP. © 2017 Elsevier Inc. All rights reserved.

Multi-Scale Feature Pyramid for Detection of Red Lesions in Fundus Images

Diabetic retinopathy (DR) is increasing rapidly around the world, but there is a shortage of experienced ophthalmologists. Therefore, computer-based diagnosis of the fundus images is essential to screening of referable DR. Automated detection of red lesions is very important for screening of DR. This paper deals with a novel method for automatic detection of red lesion. The main contribution is developing a deep learning based detection framework to handle severe class imbalance and imbalance in sizes of red lesions. The multi-scale features are extracted using the feature pyramid network. A pyramid of features is generated with strong semantics. The proposed network is end-to-end trainable in image level with several scales and works for a wide range of red lesions with acceptable performance. Sensitivity of the proposed method is 0.76 with six false-positive per image on test set of publicly available DIARECTDB1 database and outperforms state-of-the-art approaches. A potential benefit with deep learning based detection framework could be used in screening programs of referable DR.

An effective torque-based method for automatic turn fault detection and turn fault severity classification in permanent magnet synchronous motor

An effective torque-based method for automatic turn fault detection and turn fault severity classification in permanent magnet synchronous motor

Balanced multi-scale target score network for ceramic tile surface defect detection

Ceramic tiles, as a prevalent building material, exhibit a wide variety of types and high demand. Traditional manual inspection methods relying on human visual observation suffer from low efficiency and unreliable accuracy. Current automated detection methods mostly rely on traditional image processing techniques for feature extraction, followed by machine learning-based classification. However, faced with the diversity of tile types and defect categories, fine-tuning and deployment processes require significant human and material resources, while detection efficiency remains limited. In this study, we first construct a high-resolution dataset for studying surface defects in ceramic tiles (CT surface defects dataset), encompassing multiple batches and various patterns of tiles. Subsequently, data analysis is conducted to address the scale and quantity differences in defect distribution. We propose an improved approach by introducing a content-aware feature recombination method and a dynamic attention mechanism to enhance the classical single-stage object detection algorithm YOLOv5. These enhancements aim to reduce information loss in features and enhance the expression of multi-scale features. Furthermore, we design a loss function that mitigates score differences for multi-scale defects. The proposed approach mitigates the discrepancy in contribution among different scale targets caused by imbalanced quantities. It effectively prevents the model from excessively favoring a specific scale target during the learning process. Experimental results demonstrate the superior accuracy and efficiency of our detection method.Compared to the baseline network YOLOv5, our approach achieved improvements of 4.9% in AP (Average Precision), 6% in APs (small-scale objects), and 8% in APl (large-scale objects). Furthermore, we achieved a 3.9% improvement in detecting white point defects, which are most affected by small-scale objects, and a 4.1% improvement in detecting discolored spot defect, which are most affected by class imbalance.

A Computer-Assisted Diagnostic Method for Accurate Detection of Early Nondisplaced Fractures of the Femoral Neck.

Nondisplaced femoral neck fractures are sometimes misdiagnosed by radiographs, which may deteriorate into displaced fractures. However, few efficient artificial intelligent methods have been reported. We developed an automatic detection method using deep learning networks to pinpoint femoral neck fractures on radiographs to assist physicians in making an accurate diagnosis in the first place. Our proposed accurate automatic detection method, called the direction-aware fracture-detection network (DAFDNet), consists of two steps, namely region-of-interest (ROI) segmentation and fracture detection. The first step removes the noise region and pinpoints the femoral neck region. The fracture-detection step uses a direction-aware deep learning algorithm to mark the exact femoral neck fracture location in the region detected in the first step. A total of 3840 femoral neck parts in anterior-posterior (AP) pelvis radiographs collected from the China Medical University Hospital database were used to test our method. The simulation results showed that DAFDNet outperformed the U-Net and DenseNet methods in terms of the IOU value, Dice value, and Jaccard value. Our proposed DAFDNet demonstrated over 94.8% accuracy in differentiating non-displaced Garden type I and type II femoral neck fracture cases. Our DAFDNet method outperformed the diagnostic accuracy of general practitioners and orthopedic surgeons in accurately locating Garden type I and type II fracture locations. This study can determine the feasibility of applying artificial intelligence in a clinical setting and how the use of deep learning networks assists physicians in improving correct diagnoses compared to the current traditional orthopedic manual assessments.

Determination of symmetry plane on human brain CT images

A method of automatic detection of human brain symmetry plain in computed tomography (CT) images is described. An algorithm consists of four steps: region of interest, symmetry curve detection, symmetry axis evaluation and calculation of symmetry plane. Symmetry plane are calculated using vertex slices of human brain CT images. There are approximately 40 which has symmetry curve `` ` Falx cerebri'' property. In the first, using this feature symmetry axis in slice detected. In second, coordinates of detected symmetry axis and least square method are used to calculate symmetry plane. In others images where symmetry property is not available symmetry axis is projection of calculated plane.

Detection of large-scale Spartina alterniflora removal in coastal wetlands based on Sentinel-2 and Landsat 8 imagery on Google Earth Engine

The invasion of Spartina alterniflora has posed significant threats to the ecosystem health and biodiversity in coastal wetlands in China. In recent years, China has enacted large-scale S. alterniflora removal projects. Accurate monitoring of the S. alterniflora removal dynamics is crucial for evaluating the effectiveness of the removal projects and coastal wetland management. In this study, we presented a novel method for automatic and rapid detection of S. alterniflora removal events and removal timing at large scale based on time series Sentinel-2 and Landsat 8 imagery on Google Earth Engine. The method first detected and reconstructed the tidal-inundation-related Normalized Difference Vegetation Index (NDVI) using a newly proposed Tide Gap Filling algorithm, aiming to alleviate the impact of tidal inundation on S. alterniflora removal detection; then, the potential removal period were extracted based on the reconstructed NDVI time series; finally, the removal event and the corresponding removal timing were identified and mapped within the potential removal period by considering the phenological characteristics of the undisturbed S. alterniflora. We took the Shandong Province in northern coastal China and Fujian Province in southern coastal China as our study areas, where large-scale S. alterniflora removal projects were carried out in 2021 and 2022, respectively. Accuracy assessment based on field surveys and high-spatial-resolution PlanetScope imagery showed that the method achieved good accuracies for removal detection, with overall accuracies of 88.89 % for Shandong and 81.33 % for Fujian, respectively. The mean absolute error day (MAED) between the detected removal timing and the reference removal timing was 6.77 days in Shandong and 13.75 days in Fujian. Both provinces have successfully implemented the S. alterniflora removal projects. In Shandong, 82.32 % (7,243.94 ha) of S. alterniflora were removed during July to December in 2021; in Fujian, 88.81 % (6,649.21 ha) of S. alterniflora were removed during July to December in 2022. We expect that the method has great potential to be applied at national scale to provide baseline data for coastal wetland restoration and management.

Automated detection of Crop-Row lines and measurement of maize width for boom spraying

Crop-row line detection is key to realizing the accurate spraying of crops; however, it is challenging because of uneven lighting, weeds, and missing plants within rows. In this study, we propose an automatic detection method for maize width and row centerlines based on image processing that can detect maize at various growth stages and under complex field conditions. Our method consists of three stages: (i) Maize crop-row segmentation based on ResC-UNet, an improved version of U-Net that integrates ResNet-50 and attention mechanism modules. (ii) Correction of distorted images by adaptive perspective transformation to prevent parallel crop rows from creating “vanishing points” at the top of the image. (iii) Detection of the width and missing plants in the row through width fitting, and extraction of the actual row line according to the positional relationship between the width and row line. We established a dataset of maize crop rows by capturing images of maize at different growth stages in the field. Offline experiments on this dataset showed that our method achieved 94.9% accuracy for maize row segmentation, 1.71° average error angle for row-line fitting, and 91.51% average pixel accuracy for width detection. It thus enables technical support for precise variable-rate spraying along rows during the maize seedling stage.

Automatic Detection and Segmentation of Brain Hemorrhage Based on Improved U-Net Model.

Brain hemorrhage is one of the leading causes of death due to the sudden rupture of a blood vessel in the brain, resulting in bleeding in the brain parenchyma. The early detection and segmentation of brain damage are extremely important for prompt treatment. Some previous studies focused on localizing cerebral hemorrhage based on bounding boxes without specifying specific damage regions. However, in practice, doctors need to detect and segment the hemorrhage area more accurately. In this paper, we propose a method for automatic brain hemorrhage detection and segmentation using the proposed network models, which are improved from the U-Net by changing its backbone with typical feature extraction networks, i.e., DenseNet-121, ResNet-50, and MobileNet-V2. The U-Net architecture has many outstanding advantages. It does not need to do too many preprocessing techniques on the original images and it can be trained with a small dataset providing low error segmentation in medical images. We use the transfer learning approach with the head CT dataset gathered on Kaggle including two classes, bleeding and non-bleeding. Besides, we give some comparison results between the proposed models and the previous works to provide an overview of the suitable model for cerebral CT images. On the head CT dataset, our proposed models achieve a segmentation accuracy of up to 99%.

Automated detection of through-cracks in pavement using three-instantaneous attributes fusion and Swin Transformer network

To improve the performance of existing through-crack detection networks by solving the problem in which through-cracks are misidentified as simple surface cracks due to limited feature extraction, this study proposes an automated detection method based on the fusion of three instantaneous attributes and the Swin Transformer network. First, a 900MHZ ground-coupled radar system was used to collect data and construct the original dataset (Origin). Then, the Hilbert-Huang transform was used to extract three instantaneous attributes (instantaneous amplitude (IA), instantaneous phase (IP) and instantaneous frequency (IF)). Second, four fusion-feature datasets, i.e., IA + IP, IA + IF, IP + IF and IA + IP + IF, were constructed using the spectral weighting of individual features. Finally, the Swin Transformer network was proposed to detect through-cracks. The results show that the IA + IF dataset exhibited the best performance. The improved network achieved a 5.4% increase in the mean average precision compared with the initial network, reaching 87.78%.

Improved STMask R-CNN-based defect detection model for automatic visual inspection of an optics lens.

A lens defect is a common quality issue that has seriously harmed the scattering characteristics and performance of optical elements, reducing the quality consistency of the finished products. Furthermore, the energy hotspots coming from the high-energy laser through diffraction of optical component defects are amplified step by step in multi-level laser conduction, causing serious damage to the optical system. Traditional manual detection mainly relies on experienced workers under a special light source environment with high labor intensity, low efficiency, and accuracy. The common machine vision techniques are incapable of detecting low contrast and complex morphological defects. To address these challenges, a deep learning-based method, named STMask R-CNN, is proposed to detect defects on the surface and inside of a lens in complex environments. A Swin Transformer, which focuses on improving the modeling and representation capability of the features in order to improve the detection performance, is incorporated into the Mask R-CNN in this case. A challenge dataset containing more than 3800 images (18000 defect sample targets) with five different types of optical lens defects was created to verify the proposed approach. According to our experiments, the presented STMask R-CNN reached a precision value of 98.2%, recall value of 97.7%, F1 score of 97.9%, mAP@0.5 value of 98.1%, and FPS value of 24 f/s, which outperformed the SSD, Faster R-CNN, and YOLOv5. The experimental results demonstrated that the proposed STMask R-CNN outperformed other popular methods for multiscale targets, low contrast target detection and nesting, stacking, and intersecting defects sample detection, exhibiting good generalizability and robustness, as well as detection speed to meet mechanical equipment production efficiency requirements. In general, this research offers a favorable deep learning-based method for real-time automatic detection of optical lens defects.

Cross-patient automatic epileptic seizure detection using patient-adversarial neural networks with spatio-temporal EEG augmentation

Cross-patient automatic epileptic seizure detection through electroencephalogram (EEG) is significant for clinical application and research. However, most automatic seizure detection methods are patient-specific and have poor generalization ability to unseen patients. In this paper, we consider two aspects for solving the cross-patient generalization problem in deep learning approaches. Firstly, we propose a new data augmentation method to largely improve generalization. We analyze the statistical distribution of seizure EEG signals and propose spatio-temporal EEG augmentation (STEA) to generate synthetic training seizure data with spatio-temporal dependencies in EEG. Secondly, we propose patient-adversarial neural network (PANN) to learn a patient-invariant representation for better generalization. We add a discriminator to distinguish the identity of patients and perform adversarial optimization between the feature extractor and identity discriminator, so that only shared seizure features are maintained. We conduct experiments on both public and clinical EEG datasets with various settings of different window lengths and divisions of training and testing sets. Our model significantly improves the performance of cross-patient detection compared with other methods and achieves state-of-the-art performance. Particularly, our proposed method achieves up to 95% sensitivity on CHB-MIT under 5 s window length segments, and achieves about 85% AUC and sensitivity even on the clinical dataset. Our designed methods can enable detection models extended to unseen patients, with performance achieving the clinical application standards.

Automatic shoreline detection by processing planview timex images using bi-LSTM networks

A new automatic shoreline detection method by using a bidirectional Long Short-Term Memory (bi-LSTM) Network that processes images column by column is presented. The model is trained on manually extracted shorelines from time-exposure video-images and is very robust against the selection of images for training. Thanks to the novelty of working with image columns, instead of with the whole image, the amount of labelled images for training is limited to a few tens or even less if the conditions are good. Moreover, this column approach makes the model to be robust to variable illuminated images and more easily interpretable, light and fast. There is a wide range of configuration parameters for the bi-LSTM layer by which the system works correctly, which facilitate to use the same network in different video stations. The highest accuracy is obtained by using CIELAB colour space. Without pre-processing the raw colour channels or defining a region of interest and without post-processing the obtained shorelines, the model demonstrates impressive accuracy with mean errors of 2.8 pixels (1.4 m) in Castelldefels and 1.7 pixels (0.85 m) in Barcelona. The method could also be effective for satellite shoreline detection by using as input channel the water index of the satellite detection techniques.

Accurate apnea and hypopnea localization in PSG with Multi-scale object detection via Dual-modal Feature Learning

Localization of sleep apnea and hypopnea (SAH) events has routinely relied on expert visual inspection of polysomnography (PSG) recordings, which is a tedious task demanding a high level of professional skills. Automated detection methods have achieved remarkable success, especially with the recent advances in machine learning and deep learning technologies. However, a significant challenge still remains in methods towards clinical practices: How to accurately discriminate SAH events in PSG, with the onset and duration of each?This study develops an object detection framework for accurately identifying the position of SAH segments with varied durations (namely SAH-MOD) in three phases: (1) Dual-modal Feature Learning (DFL, dual-branch 1-D convolutional layers followed by Concatenate Block). Deep features are efficiently learned and then fused from two different types of signals related to respiratory, i.e., nasal airflow and abdominal movement; (2) Feature Map Generation (FMG, cascade 1-D convolutional layers). Feature maps are generated with multi-scale hierarchical features in different depths of the network, catering for the needs of object (SAH event) detection; Default anchors associated with the scales and receptive fields are tiled onto the corresponding detection feature maps; and (3) Multi-scale Object Detection (MOD). A variety of instances of prediction are then made on all available detection layers with post-processing to accurately capture each SAH event.Experiments have been performed on the dataset of stroke unit recordings for the detection of Obstructive Sleep Apnea Syndrome (OSASUD-dataset) with SAH-MOD against the state-of-the-art counterparts, and results indicate that: (1) SAH-MOD performs the best with a Recall of 81.0% and an F1-score of 71.1%; and (2) it has significant advantages in localizing the onset and duration of each SAH event, with 91.9% of the IoU values between predicted and labeled events falling between 0.6 and 1.0. Ablation experiments show that the introduction of dual-modal feature learning and hierarchical feature maps improves recall performance by 6.9% and 4.1%, respectively.

Aero-engine pipe gap automatic detection based on 3D scanning point clouds

Complex pipe systems are an important part of aero-engines to ensure reliable operations. These pipes are typically distributed in a narrow space while retaining a safe distance from each other. Thus, the gap between pipes needs to be periodically measured. Due to the complex structure of aero-engine pipes, there is no current effective method to detect pipe gaps. This paper proposes an automatic detection method for aero-engine pipe gaps based on raw point clouds. First, an iterative random point method is presented to obtain the center point set by relying on the geometric characteristics of the pipe. However, due to the complex background, the center point set contains significant noise and outliers. The key observation shows that the point density along a center curve and the mass of this curve are higher. We propose a mass and density-aware curve fitting algorithm capable of extracting the aero-engine pipe center curves with topological relationships and geometric details. We formulate the aero-engine pipe gap as the optimization model based on the distances between the center curves. Various experiments demonstrate the effectiveness and efficiency of the proposed approach, which exhibits superiority over several state-of-the-art methods in terms of the robustness to clutter and occlusion.

Change detection in VHR Remote Sensing Images by automatic sample selection and progressive classification

ABSTRACT This paper proposes an automatic method for land cover change detection in very-high-spatial-resolution optical remote sensing images based on the automatic selection of training samples using expectation-maximization (EM) and an extreme learning machine classifier, which has two key characteristics: (1) combining the advantages of supervised and unsupervised methods with progressive mask classification. (2) training samples of three classes (changed, unchanged, and unknown) were automatically selected by K-Means and EM, and then further refined by the likelihood function. The method was validated by one dataset of SuperView-1 imagery with a spatial resolution of 2.0 m, two datasets of TripleSat-2 imagery with a spatial resolution of 3.2 m, and one open dataset of Zi-Yuan-3 imagery with a spatial resolution of 5.8 m, and the results were compared with that of three unsupervised methods (iterative slow feature analysis, multivariate alteration detection, and adaptive object-oriented spatial-contextual extraction algorithm), two deep learning methods (convolutional-wavelet neural networks and dual-domain networks), and three supervised classifiers (support vector machine, random forest, and Naive Bayes), showing the effectiveness of this method in decrease of false-positive rates and increase of change detection accuracy. The average and maximum of the accuracy metric F1 score of our method are 0.6842 and 0.8708, respectively; the average F1 score of the unsupervised, deep learning, and supervised methods are 0.5049, 0.4943, and 0.633, respectively.

Euclid: Identification of asteroid streaks in simulated images using deep learning

The material composition of asteroids is an essential piece of knowledge in the quest to understand the formation and evolution of the Solar System. Visual to near-infrared spectra or multiband photometry is required to constrain the material composition of asteroids, but we currently have such data, especially in the near-infrared wavelengths, for only a limited number of asteroids. This is a significant limitation considering the complex orbital structures of the asteroid populations. Up to 150 000 asteroids will be visible in the images of the upcoming ESA Euclid space telescope, and the instruments of Euclid will offer multiband visual to near-infrared photometry and slitless near-infrared spectra of these objects. Most of the asteroids will appear as streaks in the images. Due to the large number of images and asteroids, automated detection methods are needed. A non-machine-learning approach based on the Streak Det software was previously tested, but the results were not optimal for short and/or faint streaks. We set out to improve the capability to detect asteroid streaks in Euclid images by using deep learning. We built, trained, and tested a three-step machine-learning pipeline with simulated Euclid images. First, a convolutional neural network (CNN) detected streaks and their coordinates in full images, aiming to maximize the completeness (recall) of detections. Then, a recurrent neural network (RNN) merged snippets of long streaks detected in several parts by the CNN. Lastly, gradient-boosted trees (XGBoost) linked detected streaks between different Euclid exposures to reduce the number of false positives and improve the purity (precision) of the sample. The deep-learning pipeline surpasses the completeness and reaches a similar level of purity of a non-machine-learning pipeline based on the StreakDet software. Additionally, the deep-learning pipeline can detect asteroids 0.25–0.5 magnitudes fainter than StreakDet. The deep-learning pipeline could result in a 50% increase in the number of detected asteroids compared to the StreakDet software. There is still scope for further refinement, particularly in improving the accuracy of streak coordinates and enhancing the completeness of the final stage of the pipeline, which involves linking detections across multiple exposures.