Map Scores Research Articles

HLG-YOLOv7: small object detection in conveyor belt damage based on leveraging hybrid local and global features

In the industrial production process, the detection of conveyor belt damage plays a crucial role in ensuring the stable operation of the transportation system. To tackle the issues of significant changes in damage size, missed detections, and poor detection ability of small-size objects in conveyor belt surface damage detection, an improved HLG-YOLOv7 (Hybrid Local and Global Features Network) conveyor belt surface defect detection algorithm is proposed. Firstly, Next-VIT is employed as the backbone network to fully extract local and global features of the damage, enhancing the model’s ability to extract features of different-sized damages. Additionally, to deeply utilize the extracted local and global features, the Explicit Visual Center (EVC) feature fusion module is introduced to obtain comprehensive and discriminative feature representations, further enhancing the detection capability of small objects. Lastly, a lightweight neck structure is designed using GSConv to reduce the complexity of the model. Experimental results demonstrate that the proposed method has better small object detection performance compared to existing methods. The improved algorithm achieves mAP and F1 scores of 96.24% and 97.15%, respectively, with an FPS of 28.2.

DNN-PolSAR: Urban Image Segmentation and Classification using Polarimetric SAR based on DNNs

Synthetic Aperture Radar (SAR) image segmentation and classification is a popular technique for learning and detection of objects such as buildings, trees, monuments, crops water-bodies, hills, etc. SAR technique is being used for urban development and city-planning, building control of municipal objects, searching best locations, detection of changes in the existing systems, etc. using polarimetry based on Deep Neural Networks. In this paper, weproposed a technique for Urban Image Segmentation and Classification using Polarimetric SAR based on Deep NeuralNetworks (DNN-PolSAR). In our proposed DNN-PolSAR technique, we useMask-RCNN, LinkNet, FPN, and PSP- Net as model architectures, whereas ResNet50, ResNet101, ResNet152, and VGG-19 are used as backbone networks.We first apply polarimetric decomposition on airborne Uninhabited Aerial Vehicle Synthetic Aperture (UAVSAR) im- ages of urban areas and then the decomposed images are fed to DNNs for segmentation and classification. We then simulate DNN-PolSAR considering different hyper-parameters and compare the obtained scores of hyper-parametersagainst used model architectures and backbone networks. In comparison, it is found that DNN-PolSAR based on FPN model with ResNet152 performed the best for segmentation and classification. The mean Average Precision (mAP) score of the DNN-PolSAR based on FPN with a pixel accuracy of 90.9% is 0.823, which outperforms other Deep Learning models.

MILITARY-EYES: A Real-Time Hyper-Tuned and Optimised Object Detection Approach Tailored for Military Dataset

Accurately identifying all objects of interest inside the specified frame of reference is crucial for object identification techniques to allow machine vision to interpret images successfully. Theoretical frameworks from computer vision and deep learning have informed many potential solutions to this problem. However, current approaches frequently fail when faced with objects going through random geometric changes and continually show shortcomings when identifying small, dense objects. This research looks at state-of-the-art object detection methods, compares them, and then suggests a convolutional object detection network that has been tweaked to fix the problems with the existing methods. Comparing our research to existing approaches, we find that they perform better. We accomplish this by training deep convolutional networks to detect geometric transformations and adjusting the networks to handle multi-scaled features. The results achieved after the optimization of the You only look once (YOLO) V5 & V7 are stated in this paper. The experiments demonstrated that YOLO v8 achieved higher accuracy than the hyper-tuned YOLO V5 and V7, with an average Precision (mAP) score of 52.8% on the customized military dataset running at 60 Frames Per Second (FPS). Object detection using deep neural networks (DNNs) poses a significant challenge due to the high computational and power requirements, mainly when deployed on general-purpose platforms such as CPUs and GPUs. However, addressing these limitations becomes crucial for efficient edge computing tasks like object detection, where faster, smaller, and energy-efficient solutions are essential. To overcome these challenges, system-on-chip (SoC) designs emerge as a promising solution.

HVConv: Horizontal and Vertical Convolution for Remote Sensing Object Detection

Generally, the interesting objects in aerial images are completely different from objects in nature, and the remote sensing objects in particular tend to be more distinctive in aspect ratio. The existing convolutional networks have equal aspect ratios of the receptive fields, which leads to receptive fields either containing non-relevant information or being unable to fully cover the entire object. To this end, we propose Horizontal and Vertical Convolution, which is a plug-and-play module to address different aspect ratio problems. In our method, we introduce horizontal convolution and vertical convolution to expand the receptive fields in the horizontal and vertical directions, respectively, to reduce redundant receptive fields, so that remote sensing objects with different aspect ratios can achieve better receptive fields coverage, thereby achieving more accurate feature representation. In addition, we design an attention module to dynamically aggregate these two sub-modules to achieve more accurate feature coverage. Extensive experimental results on the DOTA and HRSC2016 datasets show that our HVConv achieves accuracy improvements in diverse detection architectures and obtains SOTA accuracy (mAP score of 77.60% with DOTA single-scale training and mAP score of 81.07% with DOTA multi-scale training). Various ablation studies were conducted as well, which is enough to verify the effectiveness of our model.

Interrelatedness patterns of knowledge representation in extension concept mapping

Extension concept mapping enhances existing maps by integrating new knowledge, yielding an additional map. This study focuses on two potential extension designs: Extended Kit-Build and Extended Scratch-Build methods. While prior research favored Extended Kit-Build for cognitive knowledge comprehension and map scores, it lacked insights into concept map-relatedness patterns. The tight connections among knowledge representations demonstrate a remarkable level of expertise and the profoundness of an individual’s understanding. This study reveals interrelatedness patterns in extension concept mapping activities that connect previous and new knowledge. The dataset was obtained from a database that accommodates the results of concept mapping activities of 55 university students on two material topics. The study employed a two-group design, wherein the experimental cohort embraced the Extended Kit-Build approach, contrasting with the control cohort’s utilization of the Extended Scratch-Build approach. Extension Relationship scores were utilized to evaluate the knowledge interrelatedness patterns in extension concept mapping. The scoring method assessed both the number and quality of concept map proposition relationships. The experimental group established a statistically more significant quantity and qualitative strength of extension relationships than those within the control group. In the experimental group, a statistically noteworthy positive correlation emerged between the scores of extension relationships and students’ comprehension.

Lightweight-Improved YOLOv5s Model for Grape Fruit and Stem Recognition

Mechanized harvesting is the key technology to solving the high cost and low efficiency of manual harvesting, and the key to realizing mechanized harvesting lies in the accurate and fast identification and localization of targets. In this paper, a lightweight YOLOv5s model is improved for efficiently identifying grape fruits and stems. On the one hand, it improves the CSP module in YOLOv5s using the Ghost module, reducing model parameters through ghost feature maps and cost-effective linear operations. On the other hand, it replaces traditional convolutions with deep convolutions to further reduce the model’s computational load. The model is trained on datasets under different environments (normal light, low light, strong light, noise) to enhance the model’s generalization and robustness. The model is applied to the recognition of grape fruits and stems, and the experimental results show that the overall accuracy, recall rate, mAP, and F1 score of the model are 96.8%, 97.7%, 98.6%, and 97.2% respectively. The average detection time on a GPU is 4.5 ms, with a frame rate of 221 FPS, and the weight size generated during training is 5.8 MB. Compared to the original YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x models under the specific orchard environment of a grape greenhouse, the proposed model improves accuracy by 1%, decreases the recall rate by 0.2%, increases the F1 score by 0.4%, and maintains the same mAP. In terms of weight size, it is reduced by 61.1% compared to the original model, and is only 1.8% and 5.5% of the Faster-RCNN and SSD models, respectively. The FPS is increased by 43.5% compared to the original model, and is 11.05 times and 8.84 times that of the Faster-RCNN and SSD models, respectively. On a CPU, the average detection time is 23.9 ms, with a frame rate of 41.9 FPS, representing a 31% improvement over the original model. The test results demonstrate that the lightweight-improved YOLOv5s model proposed in the study, while maintaining accuracy, significantly reduces the model size, enhances recognition speed, and can provide fast and accurate identification and localization for robotic harvesting.

Proactive hybrid learning framework for real-time multi-vehicle detection in unregulated traffic environments

Reliable multi-vehicle detection in unregulated traffic environments is a crucial computer vision task in the development of Intelligent Transportation Systems (ITS). Despite the promising potential of Deep Learning (DL) methods for vehicle detection, the presence of uncertainties such as varying vehicle shapes and sizes, intricate background clutter, and unpredictable vehicle flow contribute to the chaotic unregulated traffic. For real-time vehicle detection in an unregulated traffic environment this paper proposes the Hybrid Learning Multi-Vehicle Detection framework (HL-MVD) based on the Convolutional Multi-head Attention Transformer Detector (CMATDet). The primary objective of this study is to generate a dataset containing highly informative video frames using a Pool-based Active Learning Strategy (PALS). Additionally, transfer learning will be implemented to train CMATDet to achieve improved accuracy and reduced detection latency. The proposed approach restructures the baseline YOLOv5x and incorporates a Multi-head Attention transformer encoder to effectively extract global features and a Scale Specific Bidirectional Feature Pyramid Network (SS-BiFPN) to facilitate multi-scale feature representation. Simplified Optimal Transport Algorithm with top-q approximation technique (Sim-OTA) is utilized for label assignment approach. Heatmap analysis demonstrated the suitability of the newly generated dataset “AU-INV-P-PALS” for detecting specific Indian native vehicles. The performance of the proposed framework is evaluated on our custom-developed AU-INV-P-PALS vehicle dataset and the IITM-HeTra Dataset 1. In comparison with contemporary detection models, the proposed HL-MVD framework resulted with higher mAP scores (91.1% on mAP@0.5 and 78.3% on mAP@0.5:0.95) for AU-INV-P-PALS. The proposed model demonstrated lower inference latency (8.1 ms), higher precision score (82.7% for IoU = 0.5), and higher recall score (90.8% for IoU = 0.5) than recent deep learning-based detection models in the literature. The top-q approximation technique in the detection head results in a reduced false-positive rate compared to conventional models. Finally, the performance of the proposed framework is tested on CCTV traffic footage captured on city roads in Chennai, India.

An open flame and smoke detection dataset for deep learning in remote sensing based fire detection

ABSTRACT The early warning of fires is pivotal for preventing substantial economic losses and ecological damage. However, the enhancement of fire model generalization performance still faces challenges due to limitations in existing fire datasets, such as image quantity and heterogeneity. Taking the advantage of rapid advancements in remote sensing technology and artificial intelligence, we introduce the Flame And Smoke Detection Dataset (FASDD), a pioneering collection comprising over 120,000 heterogeneous images covering various fire scenarios, aiming to promote the evolution of fire detection models. FASDD serves as a challenging benchmark for object detection tasks and features three sub-datasets: FASDD_CV, FASDD_UAV, and FASDD_RS. These sub-datasets comprise cross-domain images from ground-based, airborne, and spaceborne sensors, respectively. Extensive experiments using advanced Swin Transformer models demonstrate satisfactory fire detection performance, with mAP scores of 84.9%, 89.7%, and 74.0% on the respective sub-datasets. Notably, the application case in wildfire localization highlights the proficiency of FASDD_RS models for monitoring and tracking forest fires. Trained on FASDD_RS, FASDD_UAV, and FASDD_CV, deep learning models can be individually deployed on edge devices including satellites, drones, and terrestrial sensors, thus empowering collaborative fire warning within the space-air-ground integrated observation network. Additionally, a standardized procedure for data preparation, data procedure, and quality assurance is established, providing off-the-shelf annotations with four common configurations to facilitate machine learning model training. FASDD stands as a pivotal benchmark, providing valuable resources for training and evaluating deep learning fire detection models, especially Large Vision Models (LVM). It holds the potential to accelerate research progress in urban firefighting or forest fire detection, providing vital security measures for early fire warning and emergency response tasks.

A Comparative Dataset of Annotated Broccoli Heads Recorded with Depth Cameras from a Moving Vehicle

An extensive, publicly available dataset is presented—the LAR Broccoli dataset—which contains 20,000 manually annotated images of broccoli heads captured from a moving tractor at an organic farm in the UK. The dataset contains images of the same row of broccoli heads recorded at 30 frames per second (fps) with three different cameras. Two off-the-shelf, relatively low-cost depth-sensing cameras were used, with the tractor moving at a speed of around 1 km/h, in addition to a webcam, with the tractor moving twice as fast. The utility of the dataset is demonstrated in four ways. First, three different state-of-the-art detector models were trained on the dataset, achieving an overall mean Average Precision (mAP) score of over 95% for the best-performing detector. The results validate the utility of the dataset for the standard task of in-field broccoli head recognition. Second, experiments with transfer learning were conducted, initialised with a smaller pre-trained broccoli detection model, and refined with the LAR Broccoli dataset. Third, we assessed the advantages of transfer learning not only using mAP but also according to time and space requirements for training models, which provides a proxy metric for energy efficiency, a practical consideration for real-world model training. Fourth, the cross-camera generalisation among the three camera systems was compared. The results highlight that testing and training detector models using different camera systems can lead to reduced performance, unless the training set also includes some images captured in the same manner as those in the test set.

Enabling near-real-time safety glove detection through edge computing and transfer learning: comparative analysis of edge and cloud computing-based methods

PurposeComputer vision and deep learning (DL) methods have been investigated for personal protective equipment (PPE) monitoring and detection for construction workers’ safety. However, it is still challenging to implement automated safety monitoring methods in near real time or in a time-efficient manner in real construction practices. Therefore, this study developed a novel solution to enhance the time efficiency to achieve near-real-time safety glove detection and meanwhile preserve data privacy.Design/methodology/approachThe developed method comprises two primary components: (1) transfer learning methods to detect safety gloves and (2) edge computing to improve time efficiency and data privacy. To compare the developed edge computing-based method with the currently widely used cloud computing-based methods, a comprehensive comparative analysis was conducted from both the implementation and theory perspectives, providing insights into the developed approach’s performance.FindingsThree DL models achieved mean average precision (mAP) scores ranging from 74.92% to 84.31% for safety glove detection. The other two methods by combining object detection and classification achieved mAP as 89.91% for hand detection and 100% for glove classification. From both implementation and theory perspectives, the edge computing-based method detected gloves faster than the cloud computing-based method. The edge computing-based method achieved a detection latency of 36%–68% shorter than the cloud computing-based method in the implementation perspective. The findings highlight edge computing’s potential for near-real-time detection with improved data privacy.Originality/valueThis study implemented and evaluated DL-based safety monitoring methods on different computing infrastructures to investigate their time efficiency. This study contributes to existing knowledge by demonstrating how edge computing can be used with DL models (without sacrificing their performance) to improve PPE-glove monitoring in a time-efficient manner as well as maintain data privacy.

One-stop detection of anterior cruciate ligament injuries on magnetic resonance imaging using deep learning with multicenter validation.

Anterior cruciate ligament (ACL) injuries are closely associated with knee osteoarthritis (OA). However, diagnosing ACL injuries based on knee magnetic resonance imaging (MRI) has been subjective and time-consuming for clinical doctors. Therefore, we aimed to devise a deep learning (DL) model leveraging MRI to enable a comprehensive and automated approach for the detection of ACL injuries. A retrospective study was performed extracting data from the Osteoarthritis Initiative (OAI). A total of 1,589 knees (comprising 1,443 intact, 90 with partial tears, and 56 with full tears) were enrolled to construct the classification model. This one-stop detection pipeline was developed using a tailored YOLOv5m architecture and a ResNet-18 convolutional neural network (CNN) to facilitate tasks based on sagittal 2-dimensional (2D) intermediate-weighted fast spin-echo sequence at 3.0T. To ensure the reliability and robustness of the classification system, it was subjected to external validation across 3 distinct datasets. The accuracy, sensitivity, specificity, and the mean average precision (mAP) were utilized as the evaluation metric for the model performance by employing a 5-fold cross-validation approach. The radiologist's interpretations were employed as the reference for conducting the evaluation. The localization model demonstrated an accuracy of 0.89 and a sensitivity of 0.93, achieving a mAP score of 0.96. The classification model demonstrated strong performance in detecting intact, partial tears, and full tears at the optimal threshold on the internal dataset, with sensitivities of 0.941, 0.833, and 0.929, specificities of 0.925, 0.947, and 0.991, and accuracies of 0.940, 0.941, and 0.989, respectively. In comparison, on a subset consisting of 171 randomly selected knees from the OAI, the radiologists demonstrated a sensitivity ranging between 0.660 and 1.000, specificity ranging between 0.691 and 1.000, and accuracy ranging between 0.689 and 1.000. On a subset consisting of 170 randomly selected knees from the Chinese dataset, the radiologists exhibited a sensitivity ranging between 0.711 and 0.948, specificity ranging between 0.768 and 0.977, and accuracy ranging between 0.683 and 0.917. After retraining, the model achieved sensitivities ranging between 0.630 and 0.961, specificities ranging between 0.860 and 0.961, and accuracies ranging between 0.832 and 0.951, respectively, on the external validation dataset. The proposed model utilizing knee MRI showcases robust performance in the domains of ACL localization and classification.

FSNet: Enhancing Forest-Fire and Smoke Detection with an Advanced UAV-Based Network

Forest fires represent a significant menace to both the ecological equilibrium of forests and the safety of human life and property. Upon ignition, fires frequently generate billowing smoke. The prompt identification and management of fire sources and smoke can efficiently avert the occurrence of extensive forest fires, thereby safeguarding both forest resources and human well-being. Although drone patrols have emerged as a primary method for forest-fire prevention, the unique characteristics of forest-fire images captured from high altitudes present challenges. These include remote distances, small fire points, smoke targets with light hues, and complex, ever-changing background environments. Consequently, traditional target-detection networks frequently exhibit diminished accuracy when handling such images. In this study, we introduce a cutting-edge drone-based network designed for the detection of forest fires and smoke, named FSNet. To begin, FSNet employs the YOCO data-augmentation method to enhance image processing, thereby augmenting both local and overall diversity within forest-fire images. Next, building upon the transformer framework, we introduce the EBblock attention module. Within this module, we introduce the notion of “groups”, maximizing the utilization of the interplay between patch tokens and groups to compute the attention map. This approach facilitates the extraction of correlations among patch tokens, between patch tokens and groups, and among groups. This approach enables the comprehensive feature extraction of fire points and smoke within the image, minimizing background interference. Across the four stages of the EBblock, we leverage a feature pyramid to integrate the outputs from each stage, thereby mitigating the loss of small target features. Simultaneously, we introduce a tailored loss function, denoted as Lforest, specifically designed for FSNet. This ensures the model’s ability to learn effectively and produce high-quality prediction boxes. We assess the performance of the FSNet model across three publicly available forest-fire datasets, utilizing mAP, Recall, and FPS as evaluation metrics. The outcomes reveal that FSNet achieves remarkable results: on the Flame, Corsican, and D-Fire datasets, it attains mAP scores of 97.2%, 87.5%, and 94.3%, respectively, with Recall rates of 93.9%, 87.3%, and 90.8%, respectively, and FPS values of 91.2, 90.7, and 92.6, respectively. Furthermore, extensive comparative and ablation experiments validate the superior performance of the FSNet model.

TANet: Text region attention learning for vehicle re-identification

In recent years, the challenge of distinguishing vehicles of the same model has prompted a shift towards leveraging both global appearances and local features, such as lighting and rearview mirrors, for vehicle re-identification (ReID). Despite advancements, accurately identifying vehicles remains complex, particularly due to the underutilization of highly discriminative text regions. This paper introduces the Text Region Attention Network (TANet), a novel approach that integrates global and local information with a specific focus on text regions for improved feature learning. TANet uniquely captures stable and distinctive features across various vehicle views, demonstrating its effectiveness through rigorous evaluation on the VeRi-776, VehicleID, and VERI-Wild datasets. TANet significantly outperforms existing methods, achieving mAP scores of 83.6% on VeRi-776, 84.4% on VehicleID (Large), and 76.6% on VERI-Wild (Large). Statistical tests further validate the superiority of TANet over the baseline, showcasing notable improvements in mAP and Top-1 through Top-15 accuracy metrics.

YOLO-Based Models for Smoke and Wildfire Detection in Ground and Aerial Images

Wildland fires negatively impact forest biodiversity and human lives. They also spread very rapidly. Early detection of smoke and fires plays a crucial role in improving the efficiency of firefighting operations. Deep learning techniques are used to detect fires and smoke. However, the different shapes, sizes, and colors of smoke and fires make their detection a challenging task. In this paper, recent YOLO-based algorithms are adopted and implemented for detecting and localizing smoke and wildfires within ground and aerial images. Notably, the YOLOv7x model achieved the best performance with an mAP (mean Average Precision) score of 80.40% and fast detection speed, outperforming the baseline models in detecting both smoke and wildfires. YOLOv8s obtained a high mAP of 98.10% in identifying and localizing only wildfire smoke. These models demonstrated their significant potential in handling challenging scenarios, including detecting small fire and smoke areas; varying fire and smoke features such as shape, size, and colors; the complexity of background, which can include diverse terrain, weather conditions, and vegetation; and addressing visual similarities among smoke, fog, and clouds and the the visual resemblances among fire, lighting, and sun glare.

RETRACTED: Utilizing Generative Adversarial Networks for Acne Dataset Generation in Dermatology

Background: In recent years, computer-aided diagnosis for skin conditions has made significant strides, primarily driven by artificial intelligence (AI) solutions. However, despite this progress, the efficiency of AI-enabled systems remains hindered by the scarcity of high-quality and large-scale datasets, primarily due to privacy concerns. Methods: This research circumvents privacy issues associated with real-world acne datasets by creating a synthetic dataset of human faces with varying acne severity levels (mild, moderate, and severe) using Generative Adversarial Networks (GANs). Further, three object detection models—YOLOv5, YOLOv8, and Detectron2—are used to evaluate the efficacy of the augmented dataset for detecting acne. Results: Integrating StyleGAN with these models, the results demonstrate the mean average precision (mAP) scores: YOLOv5: 73.5%, YOLOv8: 73.6%, and Detectron2: 37.7%. These scores surpass the mAP achieved without GANs. Conclusions: This study underscores the effectiveness of GANs in generating synthetic facial acne images and emphasizes the importance of utilizing GANs and convolutional neural network (CNN) models for accurate acne detection.

Diabetic Foot Ulcer Detection: Combining Deep Learning Models for Improved Localization

Diabetes mellitus (DM) can cause chronic foot issues and severe infections, including Diabetic Foot Ulcers (DFUs) that heal slowly due to insufficient blood flow. A recurrence of these ulcers can lead to 84% of lower limb amputations and even cause death. High-risk diabetes patients require expensive medications, regular check-ups, and proper personal hygiene to prevent DFUs, which affect 15–25% of diabetics. Accurate diagnosis, appropriate care, and prompt response can prevent amputations and fatalities through early and reliable DFU detection from image analysis. We propose a comprehensive deep learning-based system for detecting DFUs from patients’ feet images by reliably localizing ulcer points. Our method utilizes innovative model ensemble techniques—non-maximum suppression (NMS), Soft-NMS, and weighted bounding box fusion (WBF)—to combine predictions from state-of-the-art object detection models. The performances of diverse cutting-edge model architectures used in this study complement each other, leading to more generalized and improved results when combined in an ensemble. Our WBF-based approach combining YOLOv8m and FRCNN-ResNet101 achieves a mean average precision (mAP) score of 86.4% at the IoU threshold of 0.5 on the DFUC2020 dataset, significantly outperforming the former benchmark by 12.4%. We also perform external validation on the IEEE DataPort Diabetic Foot dataset which has demonstrated robust and reliable model performance on the qualitative analysis. In conclusion, our study effectively developed an innovative diabetic foot ulcer (DFU) detection system using an ensemble model of deep neural networks (DNNs). This AI-driven tool serves as an initial screening aid for medical professionals, augmenting the diagnostic process by enhancing sensitivity to potential DFU cases. While recognizing the presence of false positives, our research contributes to improving patient care through the integration of human medical expertise with AI-based solutions in DFU management.

NeXtFusion: Attention-Based Camera-Radar Fusion Network for Improved Three-Dimensional Object Detection and Tracking

Accurate perception is crucial for autonomous vehicles (AVs) to navigate safely, especially in adverse weather and lighting conditions where single-sensor networks (e.g., cameras or radar) struggle with reduced maneuverability and unrecognizable targets. Deep Camera-Radar fusion neural networks offer a promising solution for reliable AV perception under any weather and lighting conditions. Cameras provide rich semantic information, while radars act like an X-ray vision, piercing through fog and darkness. This work proposes a novel, efficient Camera-Radar fusion network called NeXtFusion for robust AV perception with an improvement in object detection accuracy and tracking. Our proposed approach of utilizing an attention module enhances crucial feature representation for object detection while minimizing information loss from multi-modal data. Extensive experiments on the challenging nuScenes dataset demonstrate NeXtFusion’s superior performance in detecting small and distant objects compared to other methods. Notably, NeXtFusion achieves the highest mAP score (0.473) on the nuScenes validation set, outperforming competitors like OFT (35.1% improvement) and MonoDIS (9.5% improvement). Additionally, NeXtFusion demonstrates strong performance in other metrics like mATE (0.449) and mAOE (0.534), highlighting its overall effectiveness in 3D object detection. Furthermore, visualizations of nuScenes data processed by NeXtFusion further demonstrate its capability to handle diverse real-world scenarios. These results suggest that NeXtFusion is a promising deep fusion network for improving AV perception and safety for autonomous driving.

Document Retrieval System for Biomedical Question Answering

In this paper, we describe our biomedical document retrieval system and answers extraction module, which is part of the biomedical question answering system. Approximately 26.5 million PubMed articles are indexed as a corpus with the Apache Lucene text search engine. Our proposed system consists of three parts. The first part is the question analysis module, which analyzes the question and enriches it with biomedical concepts related to its wording. The second part of the system is the document retrieval module. In this step, the proposed system is tested using different information retrieval models, like the Vector Space Model, Okapi BM25, and Query Likelihood. The third part is the document re-ranking module, which is responsible for re-arranging the documents retrieved in the previous step. For this study, we tested our proposed system with 6B training questions from the BioASQ challenge task. We obtained the best MAP score on the document retrieval phase when we used Query Likelihood with the Dirichlet Smoothing model. We used the sequential dependence model at the re-rank phase, but this model produced a worse MAP score than the previous phase. In similarity calculation, we included the Named Entity Recognition (NER), UMLS Concept Unique Identifiers (CUI), and UMLS Semantic Types of the words in the question to find the sentences containing the answer. Using this approach, we observed a performance enhancement of roughly 25% for the top 20 outcomes, surpassing another method employed in this study, which relies solely on textual similarity.

Causal Reinforcement Learning for Knowledge Graph Reasoning

Knowledge graph reasoning can deduce new facts and relationships, which is an important research direction of knowledge graphs. Most of the existing methods are based on end-to-end reasoning which cannot effectively use the knowledge graph, so consequently the performance of the method still needs to be improved. Therefore, we combine causal inference with reinforcement learning and propose a new framework for knowledge graph reasoning. By combining the counterfactual method in causal inference, our method can obtain more information as prior knowledge and integrate it into the control strategy in the reinforcement model. The proposed method mainly includes the steps of relationship importance identification, reinforcement learning framework design, policy network design, and the training and testing of the causal reinforcement learning model. Specifically, a prior knowledge table is first constructed to indicate which relationship is more important for the problem to be queried; secondly, designing state space, optimization, action space, state transition and reward, respectively, is undertaken; then, the standard value is set and compared with the weight value of each candidate edge, and an action strategy is selected according to the comparison result through prior knowledge or neural network; finally, the parameters of the reinforcement learning model are determined through training and testing. We used four datasets to compare our method to the baseline method and conducted ablation experiments. On dataset NELL-995 and FB15k-237, the experimental results show that the MAP scores of our method are 87.8 and 45.2, and the optimal performance is achieved.

Architecting an enterprise financial management model: leveraging multi-head attention mechanism-transformer for user information transformation

Financial management assumes a pivotal role as a fundamental information system contributing to enterprise development. Nonetheless, prevalent methodologies frequently encounter challenges in proficiently overseeing diverse information streams inherent to financial management. This study introduces an innovative paradigm for enterprise financial management centered on the transformation of user information signals. In its initial phases, the methodology augments the Transformer network and self-attention mechanism to extract features pertaining to both users and financial data, fostering a more cohesive integration of financial and user information. Subsequently, a reinforcement learning-based alignment method is implemented to reconcile disparities between financial and user information, thereby enhancing semantic alignment. Ultimately, a signal conversion technique employing generative adversarial networks is deployed to harness user information, elevating financial management efficacy and, consequently, optimizing overall financial operations. The empirical validation of this approach, achieving an impressive mAP score of 81.9%, not only outperforms existing methodologies but also underscores the tangible impact and enhanced execution prowess that this paradigm brings to financial management systems. As such, this work not only contributes to the state of the art but also holds promise for revolutionizing the landscape of enterprise financial management.