Average Precision Values Research Articles

Automatic reconstruction of semantic façade model of architectural heritage

AbstractFaçade elements, such as windows, doors, balconies, sculptures, and totems, in architectural heritage images with incomplete structures should be automatically reconstructed for applications in 3D analysis, 3D modeling, virtual tourism, city planning, and the protection and reconstruction of architectural heritage. This study segments façade elements of architectural heritage semantically using YOLOv9. A parameterized expression for the semantic façade model is designed. In addition, the façade layer graph (FLG) and element layer graph (ELG) algorithms are developed based on topological, geometric, and structural constraints to automatically reconstruct the semantic façade model for architectural heritages. The results showed that the average precision (AP) and mean intersection over union (MIoU) achieved using YOLOv9 + FLG-ELG are 86.91% and 85.63%, respectively, on the dataset concerning façade elements of architectural heritages. The AP values obtained from the proposed method are 98.5% on the ECP2011 dataset and 95.3% on the Graz2012 dataset. The YOLOv9 + FLG-ELG method automatically reconstructs regular, irregular, and complex façade layouts with high accuracy and robustness.

Can Historical Accident Data Improve Sustainable Urban Traffic Safety? A Predictive Modeling Study

Traffic safety is a critical factor for the sustainable development of urban transportation systems. This study investigates the impact of historical accident information on the prediction of future traffic accident risks, as well as the interaction between this information and other features, such as driver violations and vehicle attributes. Using a comprehensive dataset of traffic accidents involving passenger vehicles in a western Chinese city, we developed two predictive models: Model 1, which is based on vehicle information and driver violations, and Model 2, which integrates historical accident data. The results indicate that the inclusion of historical accident information significantly enhances the predictive performance of the model, particularly in terms of AUC (Area Under the Curve) and AP (Average Precision) values. Furthermore, through feature importance analysis and SHAP (SHapley Additive exPlanations) value evaluation, this study reveals the interaction effects between historical accident data and other features, and how these interactions influence model decisions. The findings suggest that historical accident data play a positive role in predicting future accident risk, with varying effects on risk mitigation. These insights provide a scientific basis for developing strategies to ensure the sustainable development of urban transportation systems.

The Development Mask R-CNN Model for Identification of Melon Plant Leaves and Branches

The quality of melons can be enhanced and optimized by pruning melon plants. Pruning is a removal process carried out on specific parts of the plant. Currently, melon plants are still pruned manually by farmers, but there are many drawbacks to this method. In this research, pruning is conducted on the branches and leaves of melon plants. Pruning can be facilitated with the assistance of a robot capable of recognizing leaves and branches. In this study, the method used to detect branches and leaves is the Mask Region-based Convolutional Neural Network (Mask R-CNN). Hyperparameter tuning technique is employed to obtain the best parameter values, including learning rate, weight decay, and learning momentum. Two scenarios are considered in this research, one with 10 epochs and the other with 30 epochs. The obtained Average Precision (AP) values at 10 epochs are 32.2% for leaf objects and 0% for branches. At 30 epochs, the AP values are 56.8% for leaf objects and 4.1% for branches. The mean Average Precision (mAP) is 16.1% for 10 epochs and 28.4% for 30 epochs.

Development of a deep learning-based object recognition system for pre-stage separation to improve the recycling rate of major general-purpose plastics

ABSTRACT The generation of plastic waste has been increasing annually, necessitating various recycling methods. Among these, material recycling with low carbon emissions should be prioritized. This study aims to enhance the material recycling rate by developing a separation system using deep learning-based object recognition. To improve labeling efficiency, image data were acquired for each material type, and the performance of different labeling methods was evaluated. The average precision (AP) values for the single-material learning model using hybrid labeling (manual + automatic) were 0.947 for PET, 0.951 for PP, 0.892 for PE, and 0.896 for PS, demonstrating superior performance compared to other methods. An integrated learning model was also developed for composite materials, achieving AP values of 0.972 for PET, 0.976 for PP, 0.963 for PE, and 0.961 for PS. These results demonstrate the model’s strong applicability for plastic waste recognition.

Enhancing Machine Learning Models Through PCA, SMOTE-ENN, and Stochastic Weighted Averaging

Predicting survival outcomes in critical accidents has been a focal point in machine learning research. This study addresses several limitations of existing methods, including insufficient management of data imbalance, lack of emphasis on hyperparameter tuning, and proneness to overfitting. Many existing models struggle to generalize effectively on imbalanced datasets or depend on default hyperparameter settings, resulting in biased predictions. By integrating Principal Component Analysis (PCA), hyperparameter optimization, and resampling methods, as well as combining Edited Nearest Neighbors (ENN) with the Synthetic Minority Oversampling Technique (SMOTE), the model significantly improves predictive accuracy and model generalization. An ensemble model combining seven machine learning algorithms—Logistic Regression, Support Vector Machine, KNN, Random Forest, XGBoost, LightGBM, and CatBoost—was applied to predict survival outcomes. Stochastic Weighted Averaging (SWA) was applied to mitigate overfitting and enhance generalization. The accuracy increased from 91.97% to 94.89% after SWA was applied in this specific scenario. The combination of PCA-based dimensionality reduction, hyperparameter tuning, and resampling techniques (ENN + SMOTE) ensured the model handled data imbalance and optimized predictive accuracy. The final model demonstrated excellent performance, with Area Under the Curve (AUC) and Average Precision (AP) values both reaching 0.98, indicating high accuracy and precision. These improvements were validated using the Titanic dataset in a binary classification problem of predicting passenger survival. The results emphasize that ensemble learning, enhanced by SWA, offers a powerful framework for handling imbalanced and complex datasets, providing significant advancements in predictive modeling accuracy. This study provides insights into how machine learning techniques can be effectively combined to solve classification challenges in real-world scenarios.

Improved YOLOv7 Electric Work Safety Belt Hook Suspension State Recognition Algorithm Based on Decoupled Head

Safety is the eternal theme of power systems. In view of problems such as time-consuming and poor real-time performance in the correct use of seat belt hooks by manual supervision operators in the process of power operation, this paper proposes an improved YOLOv7 seat belt hook suspension state recognition algorithm. Firstly, the feature extraction part of the YOLOv7 backbone network is improved, and the M-Spatial Pyramid Pooling Concurrent Spatial Pyramid Convolution (M-SPPCSPC) feature extraction module is constructed to replace the Spatial Pyramid Pooling Concurrent Spatial Pyramid Convolution (SPPCSPC) module of the backbone network, which reduces the amount of computation and improves the detection speed of the backbone network while keeping the sensory field of the backbone network unchanged. Second, a decoupled head, which realizes the confidence and regression frames separately, is introduced to alleviate the negative impact of the conflict between the classification and regression tasks, consequently improving the network detection accuracy and accelerating the network convergence. Ultimately, a dynamic non-monotonic focusing mechanism is introduced in the output layer, and the Wise Intersection over Union (WioU) loss function is used to reduce the competitiveness of high-quality anchor frames while reducing the harmful gradient generated by low-quality anchor frames, which ultimately improves the overall performance of the detection network. The experimental results show that the mean Average Precision (mAP@0.5) value of the improved network reaches 81.2%, which is 7.4% higher than that of the original YOLOv7, therefore achieving better detection results for multiple-state recognition of hooks.

YOLOv5s-Based Image Identification of Stripe Rust and Leaf Rust on Wheat at Different Growth Stages.

Stripe rust caused by Puccinia striiformis f. sp. tritici and leaf rust caused by Puccinia triticina, are two devastating diseases on wheat, which seriously affect the production safety of wheat. Timely detection and identification of the two diseases are essential for taking effective disease management measures to reduce wheat yield losses. To realize the accurate identification of wheat stripe rust and wheat leaf rust during the different growth stages, in this study, the image-based identification of wheat stripe rust and wheat leaf rust during different growth stages was investigated based on deep learning using image processing technology. Based on the YOLOv5s model, we built identification models of wheat stripe rust and wheat leaf rust during the seedling stage, stem elongation stage, booting stage, inflorescence emergence stage, anthesis stage, milk development stage, and all the growth stages. The models were tested on the different testing sets in the different individual growth stages and in all the growth stages. The results showed that the models performed differently in disease image identification. The model based on the disease images acquired during an individual growth stage was not suitable for the identification of the disease images acquired during the other individual growth stages, except for the model based on the disease images acquired during the milk development stage, which had acceptable identification performance on the testing sets in the anthesis stage and the milk development stage. In addition, the results demonstrated that wheat growth stages had a great influence on the image identification of the two diseases. The model built based on the disease images acquired in all the growth stages produced acceptable identification results. Mean F1 Score values between 64.06% and 79.98% and mean average precision (mAP) values between 66.55% and 82.80% were achieved on each testing set composed of the disease images acquired during an individual growth stage and on the testing set composed of the disease images acquired during all the growth stages. This study provides a basis for the image-based identification of wheat stripe rust and wheat leaf rust during the different growth stages, and it provides a reference for the accurate identification of other plant diseases.

Multi-Feature-Filtering-Based Road Curb Extraction from Unordered Point Clouds

Road curb extraction is a critical component of road environment perception, being essential for calculating road geometry parameters and ensuring the safe navigation of autonomous vehicles. The existing research primarily focuses on extracting curbs from ordered point clouds, which are constrained by their structure of point cloud organization, making it difficult to apply them to unordered point cloud data and making them susceptible to interference from obstacles. To overcome these limitations, a multi-feature-filtering-based method for curb extraction from unordered point clouds is proposed. This method integrates several techniques, including the grid height difference, normal vectors, clustering, an alpha-shape algorithm based on point cloud density, and the MSAC (M-Estimate Sample Consensus) algorithm for multi-frame fitting. The multi-frame fitting approach addresses the limitations of traditional single-frame methods by fitting the curb contour every five frames, ensuring more accurate contour extraction while preserving local curb features. Based on our self-developed dataset and the Toronto dataset, these methods are integrated to create a robust filter capable of accurately identifying curbs in various complex scenarios. Optimal threshold values were determined through sensitivity analysis and applied to enhance curb extraction performance under diverse conditions. Experimental results demonstrate that the proposed method accurately and comprehensively extracts curb points in different road environments, proving its effectiveness and robustness. Specifically, the average curb segmentation precision, recall, and F1 score values across scenarios A, B (intersections), C (straight road), and scenarios D and E (curved roads and ghosting) are 0.9365, 0.782, and 0.8523, respectively.

ECMTrans-net: Multi-Class Segmentation Network Based on Tumor Tissue in Endometrial Cancer Pathology Images

Endometrial cancer has the second highest incidence of malignant tumors in the female reproductive system, and accurate and efficient endometrial cancer pathology image analysis is one of the important research components of computer-aided diagnosis. However, endometrial cancer pathologic images have the challenges of smaller solid tumors, lesion areas varying in morphology, and difficulty distinguishing solid and non-solid tumors, which would impact the accuracy of subsequent pathological analyses. Therefore, an Endometrial Cancer Multi-class Transformer Network (ECMTrans-net) is proposed to improve the segmentation accuracy of endometrial cancer pathology images. On the one hand, an ECM-Attention is proposed, which can sequentially infer attention maps along three separate dimensions: channel, local spatial, and global spatial, and multiply the attention maps and the input feature map for adaptive feature refinement, solving the problems of the small size of solid tumors and similar characteristics of solid tumors to non-solid tumors and further improving the accuracy of segmentation of solid tumors. On the other hand, an ECM-Transformer is proposed, which can fuse multi-class feature information and dynamically adjust the receptive field, solving the issue of complex tumor features. Experiments on the solid tumor endometrial cancer pathological (ST-ECP) dataset show that the ECMTrans-net performs superior to state-of-the-art image segmentation methods, and the average values of Accuracy, MIoU, Precision, and Dice were 0.952, 0.927, 0.931 and 0.901, respectively.

Time-guided convolutional neural networks for spatiotemporal urban flood modelling

Urban flood modelling is key to understand flood risks and develop effective interventions in flood management. Deep learning (DL), known for its robust and automatic feature extraction capabilities, has been applied for urban flood predictions. However, the hybrid spatiotemporal structure of conventional DL-enabled urban flood models is limited in terms of accuracy and efficiency. To address this gap, this study develops a new DL model guided by time information. This model uses a classic CNN (Convolution Neural Network) architecture, Unet, as its backbone. Time information is integrated into inputs via an extra channel to specify the desired prediction time, facilitating the simulation of the spatiotemporal flood process. Additionally, a modified loss function is formulated to tackle the sample imbalance problem between flooded and non-flooded sites. The model performance is assessed in an urban area in Dalian, China with a total of 18 rainfall events of varying return periods. The model attains average precision and recall values of 0.90 and 0.81, respectively, across different time steps during various events. Furthermore, the model exhibits transferability in ungauged regions where a high influence of surrounding environments on local flood processes is identified by Grad-CAM (Gradient-weighted Class Activation Mapping) analysis. The results show that the new Unet model has great promise in efficiently providing accurate spatiotemporal flood simulations. The time-guided Unet model can serve as practical tools for rapid flood simulation in urban areas.

An efficient face and mask detection by embedded device via thermal imaging camera

Abstract Since 87.9% of COVID-19 patients develop a fever symptom, the mandatory wearing of masks, temperature screening, and isolation of individuals with fever are the most import methods for community epidemic prevention. This study develops a thermal imaging thermometer embedded system with face, face mask, and temperature detection capabilities using low-end CPU devices. The hardware includes a Raspberry Pi 4, FLIR Lepton 3.5 LWIR Micro Thermal Camera Module, display screen, and alarm. The system is characterized by its low cost and lightweight design, and it only needs to detect necessary areas around the face area during face mask and temperature detection, effectively avoiding external interference and further enhancing the effectiveness of epidemic prevention. The face detection model is based on the YOLO-Fastest architecture and has been improved, referred to as Pi-fast. In terms of performance on the training and testing sets, our proposed Pi-fast achieves Mean Average Precision (mAP) values of 99.64% and 93.18%, F1 score values of 0.94 and 0.87, and Intersection over Union (IoU)values of 78.07% and 65.98%, respectively. Regarding system performance, the frame rate of FLIR Lepton 3.5 Thermal Camera Module hardware specifications is 8.7 FPS. When using YOLO-Fastest and proposed Pi-fast model for detection on the proposed thermal imaging system, the frame rates are 6.58 FPS and 7.32FPS, respectively. The proposed Pi-fast model outperforms YOLO-Fastest about 0.74 FPS (11.2%). The system can detect objects at a distance of about 7 meters from the lens in thermographic camera with a resolution of only 160×120 pixels, making it suitable for multi-person screening and improving the screening process’s efficiency. The system can effectively perform temperature screening and face mask detection, making it suitable for office buildings, schools, medical clinics, and other public areas. During the severe epidemic period, the proposed thermal imaging embedded system that was actually used for epidemic prevention work at I-Shou University, improving screening efficiency and reducing labour costs.

An Experimental Methodology for Automated Detection of Surface Turbulence Features in Tidal Stream Environments

Tidal stream environments are important areas of marine habitat for the development of marine renewable energy (MRE) sources and as foraging hotspots for megafaunal species (seabirds and marine mammals). Hydrodynamic features can promote prey availability and foraging efficiency that influences megafaunal foraging success and behaviour, with the potential for animal interactions with MRE devices. Uncrewed aerial vehicles (UAVs) offer a novel tool for the fine-scale data collection of surface turbulence features and animals, which is not possible through other techniques, to provide information on the potential environmental impacts of anthropogenic developments. However, large imagery datasets are time-consuming to manually review and analyse. This study demonstrates an experimental methodology for the automated detection of turbulence features within UAV imagery. A deep learning architecture, specifically a Faster R-CNN model, was used to autonomously detect kolk-boils within UAV imagery of a tidal stream environment. The model was trained on pre-existing, labelled images of kolk-boils that were pre-treated using a suite of image enhancement techniques based on the environmental conditions present within each image. A 75-epoch model variant provided the highest average recall and precision values; however, it appeared to be limited by sub-optimal detections of false positive values. Although further development is required, including the creation of standardised image data pools, increased model benchmarking and the advancement of tailored pre-processing techniques, this work demonstrates the viability of utilising deep learning to automate the detection of surface turbulence features within a tidal stream environment.

SCS-YOLO: A Defect Detection Model for Cigarette Appearance

Appearance defects significantly impact cigarette quality. However, in the current high-speed production lines, manual inspection and traditional methods are unable to satisfy the actual demands of inspection. Therefore, a real-time and high-precision defect detection model for cigarette appearance, SCS-YOLO, is presented. The model integrates space-to-depth convolution (SPD-Conv), a convolutional block attention module (CBAM), and a self-calibrated convolutional module (SCConv). SPD-Conv replaces the pooling structure to enhance the granularity of feature information. CBAM improves the ability to pay attention to defect locations. Improved self-calibrated convolution broadens the network’s receptive field and feature fusion capability. Additionally, Complete IoU loss (CIoU) is replaced with Efficient IoU Loss (EIoU) to enhance model localization and mitigate sample imbalance. The experimental results show that the accuracy of SCS-YOLO is 95.5% and the mAP (mean average precision) value is 95.2%. Compared with the original model, the accuracy and mAP value of the SCS-YOLO model are improved by 4.0%. Furthermore, the model achieves a detection speed of 216 FPS, meeting cigarette production lines’ accuracy and speed demands. Our research will positively impact the real-time detection of appearance defects in cigarette production lines.

Detection of small foreign objects in Pu-erh sun-dried green tea: An enhanced YOLOv8 neural network model based on deep learning

To efficiently and accurately detect minuscule foreign objects in the processing of Pu-erh sun-dried green tea, ensuring food quality and consumer safety, this study innovatively proposes an enhanced YOLOv8 neural network based on deep learning. In light of the shortcomings of the traditional YOLOv8 network, to further enhance the model's ability to identify foreign object targets, improve the depth and breadth of feature extraction, and to better understand the contextual connections between different parts of the image, this study employs a Shape-IoU optimized loss function. It replaces parts of the network structure with Receptive-Field Attention Convolution technology and embeds Double Attention Networks to optimize the network. Experimental results show that the enhanced YOLOv8 neural network model achieves a precision rate of 98.35% for the detection of foreign objects in Pu-erh sun-dried green tea, which is a 3.93% increase compared to the original YOLOv8 network. Compared to mainstream detection models such as YOLOv7, YOLOv5, Faster-RCNN, CornerNet, and SSD, the mean Average Precision values of the enhanced YOLOv8 network model have significantly increased by 4.48%, 6.66%, 13.63%, 13.20%, and 9.84% respectively. This enhanced YOLOv8 network provides a viable research method and significant reference for the detection of small foreign objects in Pu-erh sun-dried green tea, holding substantial importance for tea-producing enterprises and food safety regulatory authorities. Furthermore, this study offers a more comprehensive and efficient solution for foreign object detection in Pu-erh tea and the broader food industry, laying the groundwork for the modernization and intelligentization of food safety and quality control.

Bayesian optimized multimodal deep hybrid learning approach for tomato leaf disease classification

Manual identification of tomato leaf diseases is a time-consuming and laborious process that may lead to inaccurate results without professional assistance. Therefore, an automated, early, and precise leaf disease recognition system is essential for farmers to ensure the quality and quantity of tomato production by providing timely interventions to mitigate disease spread. In this study, we have proposed seven robust Bayesian optimized deep hybrid learning models leveraging the synergy between deep learning and machine learning for the automated classification of ten types of tomato leaves (nine diseased and one healthy). We customized the popular Convolutional Neural Network (CNN) algorithm for automatic feature extraction due to its ability to capture spatial hierarchies of features directly from raw data and classical machine learning techniques [Random Forest (RF), XGBoost, GaussianNB (GNB), Support Vector Machines (SVM), Multinomial Logistic Regression (MLR), K-Nearest Neighbor (KNN)], and stacking for classifications. Additionally, the study incorported a Boruta feature filtering layer to capture the statistically significant features. The standard, research-oriented PlantVillage dataset was used for the performance testing, which facilitates benchmarking against prior research and enables meaningful comparisons of classification performance across different approaches. We utilized a variety of statistical classification metrics to demonstrate the robustness of our models. Using the CNN-Stacking model, this study achieved the highest classification performance among the seven hybrid models. On an unseen dataset, this model achieved average precision, recall, f1-score, mcc, and accuracy values of 98.527%, 98.533%, 98.527%, 98.525%, and 98.268%, respectively. Our study requires only 0.174 s of testing time to correctly identify noisy, blurry, and transformed images. This indicates our approach's time efficiency and generalizability in images captured under challenging lighting conditions and with complex backgrounds. Based on the comparative analysis, our approach is superior and computationally inexpensive compared to the existing studies. This work will aid in developing a smartphone app to offer farmers a real-time disease diagnosis tool and management strategies.

Histopathologic Differential Diagnosis and Estrogen Receptor/Progesterone Receptor Immunohistochemical Evaluation of Breast Carcinoma Using a Deep Learning–Based Artificial Intelligence Architecture

In breast carcinoma, invasive ductal carcinoma (IDC) is the most common histopathological subtype, and ductal carcinoma in situ (DCIS) is a precursor of IDC. They are often concomitant. The immunohistochemical staining of estrogen receptor (ER)/progesterone receptor (PR) in IDC/DCIS on whole-slide histopathological images (WSIs) can predict the prognosis of patients. However, the inter-observer variability among pathologists in reading WSIs is inevitable. Thus, artificial intelligence (AI) technology is crucial. Herein, IDC/DCIS detection was conducted by deep learning approach, including Faster R-CNN, RetinaNet, SSD300, YOLOv3, YOLOv5, YOLOv7, YOLOv8, and Swin transformer. Their performance was estimated by mean average precision (mAP) values. Cell recognition and counting were performed using AI technology to evaluate the intensity and proportion of ER/PR-immunostained cancer cells in IDC/DCIS. A three-round ring study (RS) was conducted to assess WSIs. A database for modelling the underlying probability distribution of a dataset with labels was established. YOLOv8 exhibits the highest detection performance with an mAP@0.5 of 0.944 and an mAP@0.5-0.95 of 0.790. With the assistance of YOLOv8, the scoring concordance across all pathologists was boosted to excellent in RS3 (0.970) from moderate in RS1 (0.724) and good in RS2 (0.812). Deep learning detection can be applied in clinicopathological field. To facilitate the histopathological diagnosis of IDC/DCIS and immunostaining scoring of ER/PR, a novel AI architecture and well-organized dataset were developed.

SGLMDA: A Subgraph Learning-Based Method for miRNA-Disease Association Prediction.

MicroRNAs (miRNA) are endogenous non-coding RNAs, typically around 23 nucleotides in length. Many miRNAs have been founded to play crucial roles in gene regulation though post-transcriptional repression in animals. Existing studies suggest that the dysregulation of miRNA is closely associated with many human diseases. Discovering novel associations between miRNAs and diseases is essential for advancing our understanding of disease pathogenesis at molecular level. However, experimental validation is time-consuming and expensive. To address this challenge, numerous computational methods have been proposed for predicting miRNA-disease associations. Unfortunately, most existing methods face difficulties when applied to large-scale miRNA-disease complex networks. In this paper, we present a novel subgraph learning method named SGLMDA for predicting miRNA-disease associations. For miRNA-disease pairs, SGLMDA samples K-hop subgraphs from the global heterogeneous miRNA-disease graph. It then introduces a novel subgraph representation algorithm based on Graph Neural Network (GNN) for feature extraction and prediction. Extensive experiments conducted on benchmark datasets demonstrate that SGLMDA can effectively and robustly predict potential miRNA-disease associations. Compared to other state-of-the-art methods, SGLMDA achieves superior prediction performance in terms of Area Under the Curve (AUC) and Average Precision (AP) values during 5-fold Cross-Validation (5CV) on benchmark datasets such as HMDD v2.0 and HMDD v3.2. Additionally, case studies on Colon Neoplasms and Triple-Negative Breast Cancer (TNBC) further underscore the predictive power of SGLMDA.

Intervertebral Cervical Disc Intensity (IVCDI) Detection and Classification on MRI Scans Using Deep Learning Methods

ABSTRACTRadiologists manually interpret magnetic resonance imaging (MRI) scans for the detection of intervertebral cervical disc degeneration, which are often obtained in a primary care or emergency hospital context. The ability of computer models to work with pathological findings and aid in the first interpretation of medical imaging tests is widely acknowledged. Deep learning methods, which are commonly employed today in the diagnosis or detection of many diseases, show great promise in this area. For the detection and segmentation of intervertebral cervical disc intensity, we propose a Mask‐RCNN‐based deep learning algorithm in this study. The provided approach begins by creating an original dataset using MRI scans that were collected from Yozgat Bozok University. The senior radiologist labels the data, and three classes of intensity are chosen for the classification (low, intermediate, and high). Two alternative network backbones are used in the study, and as a consequence of the training for the Mask R‐CNN algorithm, 98.14% and 96.72% mean average precision (mAP) values are obtained with the ResNet50 and ResNet101 architectures, respectively. Utilizing the five‐fold cross‐validation approach, the study is conducted. This study also applied the Faster R‐CNN method, achieving a mAP value of 85.2%. According to the author's knowledge, no study has yet been conducted to apply deep learning algorithms to detect intervertebral cervical disc intensity in a patient population with cervical intervertebral disc degeneration. By ensuring accurate MRI image interpretation and effectively supplying supplementary diagnostic information to provide accuracy and consistency in radiological diagnosis, the proposed method is proving to be a highly useful tool for radiologists.

IMAGE-BASED YOLOV4 ARCHITECTURE FOR DETECTING MINERAL IN SEDIMENTARY ROCKS THIN SECTION

Thin section analysis of sedimentary rocks is the basis for identifying minerals and textures. In general, quantitative analysis of thin sections of rock often requires many hours of work when done manually. In today's era, mineralogical interpretation and percentage calculations must be carried out automatically using more practical applications. The research method begins with the identification of 44 thin section samples in parallel plane polarized (PPL) and crossed polarized (XPL) conditions with thin section analysis then mineralogy detection is carried out using a computational approach, namely the use of image-based Deep Learning YoloV4 architecture with 2D RGB image objects from the thin section of sedimentary rock. The results of this study show the best values of Average Precision in Quartz, Feldspar, and lithic are 39.21% in the XPL model, 26.53% in the XPL model, and 15.75% in combined mode, according to the training and testing of YoloV4 Models for the identification of rock minerals in thin sections. Based on the complexity of the mineral types, the granularity of the detection, and the specific geological objectives, establishing a meaningful benchmark or baseline for comparison is always challenging. Additionally, consider discussing the trade-offs between precision and recall, as a higher precision may be more critical in some geological applications. It is expected that the application of this research can produce practical, fast and accurate interpretation of the determination of minerals in sedimentary rocks from all thin-section images of rocks and thus provide a complete understanding of geological views automatically.

DFS-DETR: Detailed-Feature-Sensitive Detector for Small Object Detection in Aerial Images Using Transformer

Object detection in aerial images plays a crucial role across diverse domains such as agriculture, environmental monitoring, and security. Aerial images present several challenges, including dense small objects, intricate backgrounds, and occlusions, necessitating robust detection algorithms. This paper addresses the critical need for accurate and efficient object detection in aerial images using a Transformer-based approach enhanced with specialized methodologies, termed DFS-DETR. The core framework leverages RT-DETR-R18, integrating the Cross Stage Partial Reparam Dilation-wise Residual Module (CSP-RDRM) to optimize feature extraction. Additionally, the introduction of the Detail-Sensitive Pyramid Network (DSPN) enhances sensitivity to local features, complemented by the Dynamic Scale Sequence Feature-Fusion Module (DSSFFM) for comprehensive multi-scale information integration. Moreover, Multi-Attention Add (MAA) is utilized to refine feature processing, which enhances the model’s capacity for understanding and representation by integrating various attention mechanisms. To improve bounding box regression, the model employs MPDIoU with normalized Wasserstein distance, which accelerates convergence. Evaluation across the VisDrone2019, AI-TOD, and NWPU VHR-10 datasets demonstrates significant improvements in the mean average precision (mAP) values: 24.1%, 24.0%, and 65.0%, respectively, surpassing RT-DETR-R18 by 2.3%, 4.8%, and 7.0%, respectively. Furthermore, the proposed method achieves real-time inference speeds. This approach can be deployed on drones to perform real-time ground detection.