检索结果-内蒙古大学图书馆

FieldNet: Efficient real-time shadow removal for enhanced vision in field robotics

EXPERT SYSTEMS WITH APPLICATIONS 2025年 279卷

作者： Saleh, Alzayat Olsen, Alex Wood, Jake Philippa, Bronson Azghadi, Mostafa Rahimi James Cook Univ Coll Sci & Engn Townsville Qld Australia AutoWeed Pty Ltd Townsville Qld Australia James Cook Univ ARC Res Hub Super charging Trop Aquaculture Genet Townsville Qld Australia

Shadows significantly hinder computer vision tasks in outdoor environments, particularly in field robotics, where varying lighting conditions complicate object detection and localization. We present FieldNet, a novel deep learning framework for real-time shadow removal, optimized for resource-constrained hardware. FieldNet introduces a probabilistic enhancement module and a novel loss function to address challenges of inconsistent shadow boundary supervision and artefact generation, achieving enhanced accuracy and simplicity without requiring shadow masks during inference. Trained on a dataset of 10,000 natural images augmented with synthetic shadows, FieldNet outperforms state-of-the-art methods on benchmark datasets (ISTD, ISTD+, SRD), with up to 9x speed improvements (66 FPS on Nvidia 2080Ti) and superior shadow removal quality (PSNR: 38.67, SSIM: 0.991). real-world case studies in precision agriculture robotics demonstrate the practical impact of FieldNet in enhancing weed detection accuracy. These advancements establish FieldNet as a robust, efficient solution for real-time vision tasks in field robotics and beyond.

关键词： Shadow removal Unpaired data real-time image processing deep learning Field robotics

来源：评论

学校读者我要写书评

暂无评论

HDMA-CGAN: Advancing image Style Transfer with deep learning

引用

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE 2024年第9期38卷

作者： Liu, Huaqun Hu, Benxi Cao, Yu Beijing Inst Graph Commun Sch New Media 1 Sect 2Xinghua St Beijing 102600 Peoples R China

The integration of artificial intelligence (AI) and deep learning heralds a transformative era in pattern recognition and computer vision, notably in image style transfer. We introduce the hierarchical dynamic multi-attention cycle generative adversarial network (HDMA-CGAN), an innovative deep learning architecture poised to redefine image style transfer capabilities. HDMA-CGAN employs a novel multi-attention mechanism and color optimization strategies, enabling precise style replication with improved fidelity and vibrancy. Our model surpasses existing benchmarks in image quality, validated by leading metrics such as peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and Fr & eacute;chet inception distance (FID). Although HDMA-CGAN advances the state of the art, it necessitates high computational resources and faces challenges with very high-resolution images. Future work could explore optimizing the model's efficiency for real-time applications and extending its application to video content. This work enhances the tools available for visual content creation and digital media enhancement, leveraging advanced pattern recognition and AI techniques to significantly impact computer vision and image processing.

关键词： deep learning image style transfer HDMA-CGAN generative adversarial networks multi-attention mechanisms color space optimization

来源：评论

学校读者我要写书评

暂无评论

Yolo-global: a real-time target detector for mineral particles

引用

JOURNAL OF real-time image processing 2024年第3期21卷 85-85页

作者： Wang, Zihao Zhou, Dong Guo, Chengjun Zhou, Ruihao Univ Elect Sci & Technol China 2006 Xiyuan Ave Chengdu Peoples R China Chongqing Changan Ind Grp Co Ltd 599 Airport Ave Chongqing Peoples R China

Recently, deep learning methodologies have achieved significant advancements in mineral automatic sorting and anomaly detection. However, the limited features of minerals transported in the form of small particles pose significant challenges to accurate detection. To address this challenge, we propose a enhanced mineral particle detection algorithm based on the YOLOv8s model. Initially, a C2f-SRU block is introduced to enable the feature extraction network to more effectively process spatial redundant information. Additionally, we designed the GFF module with the aim of enhancing information propagation between non-adjacent scale features, thereby enabling deep networks to more fully leverage spatial positional information from shallower networks. Finally, we adopted the Wise-IoU loss function to optimize the detection performance of the model. We also re-designed the position of the prediction heads to achieve precise detection of small-scale targets. The experimental results substantiate the effectiveness of the algorithm, with YOLO-Global achieving a mAP@.5 of 95.8%. In comparison to the original YOLOv8s, the improved model exhibits a 2.5% increase in mAP, achieving a model inference speed of 81 fps, meeting the requirements for real-time processing and accuracy.

关键词： Small target detection Lightweight deep learning Computer vision

来源：评论

学校读者我要写书评

暂无评论

License Plate reader with PUC Details using image processing and deep learning

License Plate reader with PUC Details using Image Processing...

引用

Smart Technologies and Systems for Next Generation Computing (ICSTSN), International Conference on

作者： R.S. Patil Sakshi Deshpande Hassan Khan Prajakta Mhatre Department of Electronics and Telecommunication Engineering Bharati Vidyapeeth College of Engineering Belpada Navi Mumbai Maharashtra India

As each vehicle is uniquely acknowledged by its license plate, the Transport System places a high priority on finding and recognizing of license plates. The news is constantly reporting on accidents and missing cars. Authorities must acknowledge all of these unlawful acts. As a result, research into the identification and recognition of vehicle number plates is ongoing. However, identifying a vehicle’s number plate has always been difficult for a number of reasons, such as brightness changes, shadows cast by moving vehicles, erratic license plate character types, different plate styles, and color effects caused by the surroundings. In this system, Number plate of vehicle is detected from a live video or an image. There is image preprocessing and segmentation done on the live video or number plate image. deep learning model methods are used, the characters from it are separated and then each character gets recognized. This helps to collect the vehicle overall project then the capabilities of different techniques into one integrated automatic system are summarized. This kind of systems can be implemented on the roadside and makes a real time comparison between passing car and list of stolen cars. This detected license plate number could also be used in car parking systems. PUC which stands for Pollution Under Control, where emission levels of vehicles and the regular renovation of the PUC certificate is done or not is verified and the details are shown. This will help in keeping an overall check on vehicles and the task which most of the places do manually to check the PUC certificate for checking status, can be verified quickly and the fine can be implemented as per so.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MDSCN: multiscale depthwise separable convolutional network for underwater graphics restoration

引用

VISUAL COMPUTER 2025年第3期41卷 1999-2010页

作者： Li, Shiyu Liu, Zehao Gao, Meijing Bai, Yang Yin, Haozheng Yanshan Univ Coll Informat Sci & Engn Key Lab Special Fiber & Fiber Sensor Hebei Prov Qinhuangdao 066004 Hebei Peoples R China Beijing Inst Technol Coll Informat & Elect Beijing 100081 Peoples R China Beijing Inst Technol Tangshan Res Inst Tangshan 063000 Peoples R China

Underwater imaging techniques have been a focus of research for computer vision. Underwater imaging frequently encounters challenges for poor image quality and slow restoration speed, thereby hindering human underwater exploration endeavors. To enhance the quality and improve the real-time performance of underwater image restoration, the paper proposes a lightweight underwater color image restoration network based on multiscale depthwise separable convolution. First, the algorithm tackles the problems of difficult convergence and slow training by improving the AdamW optimizer. Then, we propose a multiscale depthwise separable convolution module with RGB channel, which allows efficient extraction of image features based on the underwater light propagation properties. The MDSCN can effectively improve the processing speed and recovery effect of underwater images. Through experimentation and analysis, our algorithm outperforms traditional image processing methods and recent deep learning approaches in terms of visual effects and objective evaluation metrics. Furthermore, our algorithm also has a better performs than existing deep learning methods in processing speed, which demonstrates excellent generalizability and practicality. The research in the article is highly informative for the field of underwater computer vision. The dataset, training weights files and codes are publicly available https://***/raining-li/underwater-image-processing/tree/master.

关键词： Computer graphics Underwater image restoration Optimizer Multiscale convolution Depth separable convolution

来源：评论

学校读者我要写书评

暂无评论

Recent Advances in Applying Machine learning and deep learning to Detect Upper Gastrointestinal Tract Lesions

引用

IEEE ACCESS 2023年 11卷 66544-66567页

作者： Vania, Malinda Tama, Bayu Adhi Maulahela, Hasan Lim, Sunghoon Ulsan Natl Inst Sci & Technol Dept Ind Engn Ulsan 44919 South Korea Ulsan Natl Inst Sci & Technol Ind Intelligentizat Inst Ulsan 44919 South Korea Univ Maryland NSF HDR Inst Harnessing Data & Model Revolut Pola Baltimore MD 21250 USA Univ Indonesia Cipto Mangunkusumo Natl Gen Hosp Div Gastroenterol Dept Internal MedFac Med Jakarta 10430 Indonesia Ulsan Natl Inst Sci & Technol Grad Sch Artificial Intelligence Ulsan 44919 South Korea

The clinical application of a real-time artificial intelligence (AI) image processing system to diagnose upper gastrointestinal (GI) malignancies remains an experimental research and engineering problem. Understanding these commonly used technical techniques is required to appreciate the scientific quality and novelty of AI studies. Clinicians frequently lack this technical background, and AI experts may be unaware of such clinical relevance and implications in daily practice. As a result, there is a growing need for a multidisciplinary, international assessment of how to conduct high-quality AI research in upper GI malignancy detection. This research will help endoscopists build approaches or models to increase diagnosis accuracy for upper GI malignancies despite variances in experience, education, personnel, and resources, as it offers real-time and retrospective chances to improve upper GI malignancy diagnosis and screening. This comprehensive review sheds light on potential enhancements to computer-aided diagnostic (CAD) systems for GI endoscopy. The survey includes 65 studies on automatic upper GI malignancy diagnosis and evaluation, which are compared by endoscopic modalities, image counts, models, validation methods, and results. The main goal of this research is to assess and compare each AI method's current stage and potential improvement to boost performance, maturity, and the possibility to open new research areas for the application of a real-time AI image recognition system that diagnoses upper GI malignancies. The findings of this study suggest that Support Vector Machines (SVM) are frequently utilized in gastrointestinal (GI) image processing within the context of machine learning (ML). Moreover, the analysis reveals that CNN-based supervised learning object detection models are widely employed in GI image analysis within the deep learning (DL) context. The results of this study also suggest that RGB is the most commonly used image modality for GI ana

关键词： Endoscopy upper gastrointestinal tract artificial intelligence machine learning deep learning

来源：评论

学校读者我要写书评

暂无评论

TSDN: Two-Stage Raw Denoising in the Dark

引用

IEEE TRANSACTIONS ON image processing 2023年 32卷 3679-3689页

作者： Chen, Wenshu Huang, Yujie Wang, Mingyu Wu, Xiaolin Zeng, Xiaoyang Fudan Univ Coll Microelect State Key Lab ASIC & Syst Shanghai 200000 Peoples R China Shanghai ExploreX Technol Co Ltd Shanghai 200120 Peoples R China McMaster Univ Dept Elect & Comp Engn Hamilton ON L8G 4K1 Canada Shanghai ExploreX Technol Co Ltd Shanghai 200120 Peoples R China

Denoising is one of the most significant procedures in the image processing pipeline. Nowadays, deep-learning-based algorithms have achieved superior denoising quality than traditional algorithms. However, the noise becomes severe in the dark environment, where even the SOTA algorithms fail to achieve satisfactory performance. Besides, the high computational complexity of deep-learning-based denoising algorithms makes them hardware unfriendly and difficult to process high-resolution images in real-time. To address these issues, a novel low-light RAW denoising algorithm Two-Stage-Denoising (TSDN), is proposed in this paper. In TSDN, denoising consists of two procedures: noise removal and image restoration. Firstly, in the noise-removal stage, most noise is removed from the image, and an intermediate image that is easier for the network to recover the clean image is obtained. Then, in the restoration stage, the clean image is restored from the intermediate image. The TSDN is designed to be light-weight for real-time and hardware friendly. However, the tiny network will be insufficient for satisfactory performance if directly trained from scratch. Therefore, we present an Expand-Shrink-learning (ESL) method to train the TSDN. In the ESL method, firstly, the tiny network is expanded to a larger one with similar architecture but more channels and layers, which enhances the learning ability of the network because of more parameters. Secondly, the larger network is shrunk and restored to the original small network in fine-grained learning procedures, including Channel-Shrink-learning (CSL) and Layer-Shrink-learning (LSL). Experimental results demonstrate that the proposed TSDN achieves better performance (PSNR and SSIM) than other SOTA algorithms in the dark environment. Besides, the model size of TSDN is one-eighth of that of the U-Net for denoising (a classical denoising network).

关键词： image processing denoising light-weight network

来源：评论

学校读者我要写书评

暂无评论

Improving Cerebrovascular Imaging with deep learning: Semantic Segmentation for time-of-Flight Magnetic Resonance Angiography Maximum Intensity Projection image Enhancement

引用

APPLIED SCIENCES-BASEL 2025年第6期15卷 3034-3034页

作者： Yamada, Tomonari Yoshimura, Takaaki Ichikawa, Shota Sugimori, Hiroyuki Hokkaido Univ Grad Sch Hlth Sci Sapporo 0600812 Japan Hokkaido Univ Fac Hlth Sci Dept Hlth Sci & Technol Sapporo 0600812 Japan Hokkaido Univ Hosp Dept Med Phys Sapporo 0608648 Japan Hokkaido Univ Fac Med Global Ctr Biomed Sci & Engn Sapporo 0608638 Japan Hokkaido Univ Fac Med Clin AI Human Resources Dev Program Sapporo 0608638 Japan Niigata Univ Fac Med Sch Hlth Sci Dept Radiol Technol Niigata 9518518 Japan Niigata Univ Inst Res Adm Niigata 9502181 Japan Hokkaido Univ Fac Hlth Sci Dept Biomed Sci & Engn Sapporo 0600812 Japan

Magnetic Resonance Angiography (MRA) is widely used for cerebrovascular assessment, with time-of-Flight (TOF) MRA being a common non-contrast imaging technique. However, maximum intensity projection (MIP) images generated from TOF-MRA often include non-essential vascular structures such as external carotid branches, requiring manual editing for accurate visualization of intracranial arteries. This study proposes a deep learning-based semantic segmentation approach to automate the removal of these structures, enhancing MIP image clarity while reducing manual workload. Using deepLab v3+, a convolutional neural network model optimized for segmentation accuracy, the method achieved an average Dice Similarity Coefficient (DSC) of 0.9615 and an Intersection over Union (IoU) of 0.9261 across five-fold cross-validation. The developed system processed MRA datasets at an average speed of 16.61 frames per second, demonstrating real-time feasibility. A dedicated software tool was implemented to apply the segmentation model directly to DICOM images, enabling fully automated MIP image generation. While the model effectively removed most external carotid structures, further refinement is needed to improve venous structure suppression. These results indicate that deep learning can provide an efficient and reliable approach for automated cerebrovascular image processing, with potential applications in clinical workflows and neurovascular disease diagnosis.

关键词： semantic segmentation magnetic resonance angiography (MRA) deep learning

来源：评论

学校读者我要写书评

暂无评论

TraSFlow: learning traditional optical flow proposal and segmentation for optical flow estimation improvement

引用

SIGNAL image AND VIDEO processing 2025年第6期19卷 1-9页

作者： Ammar, Anis Ghozzi, Rim Souani, Chokri Sousse Univ Higher Inst Appl Sci & Technol Sousse Tunisia Fac Sci Monastir Elect & Microelect Lab Monastir Tunisia

Optical flow estimation has evolved widely, in the last decade. Several methods have been developed to achieve accurate and robust motion estimation under diverse and complex scenarios. From traditional methods to deep learning approaches, researchers aim to enhance accuracy and adaptability. This paper, introduces a novel hybrid approach that integrates deep optical flow techniques, deep segmentation and builds upon our previously proposed method for optical flow estimation. Consequently, we leverage segmentation for the optimal separation of distinct regions, and our conventional method ensures improved precision, especially in the case of simple and small movements. The execution of our proposed method, 'TrasFlow', is conducted on the Jetson Xavier NX development kit. We validate the effectiveness of our joint training program through evaluation studies adapted for the estimation of optical flow through different datasets. Our proposed method, TraSFlow, achieves significant accuracy improvement over baseline models, with an end-point error (EPE) of 4.76 in the final Sintel pass and an F1-all score of 8.23% on KITTI 2015, while maintaining real-time performance of 32 frames per second on embedded systems. These results outperform the baseline model, highlighting TrasFlow's accuracy in various scenarios and demonstrating its superiority over many state-of-the-art methods.

关键词： Motion estimation Optical flow Segmentation real-time system deepFlow

来源：评论

学校读者我要写书评

暂无评论

DSPCI-MTL: Dynamic split point computing in multi-task learning implementation with collaborative intelligence

引用

ALEXANDRIA ENGINEERING JOURNAL 2025年 124卷 404-421页

作者： Sahin, M. Faruk Yeganli, S. Faegheh Kiani, Farzad Istanbul Atlas Univ Dept Comp Engn Istanbul Turkiye Fatih Sultan Mehmet Vakif Univ Data Sci Applicat & Res Ctr VEBIM Istanbul Turkiye Univ British Columbia Dept Comp Sci Vancouver BC Canada

deep Neural Networks (DNNs) have become a crucial technology in image processing, renowned for their ability to generate effective feature maps. The integration of DNNs within Internet of Things (IoT) environments, particularly in multi-task robots and swarm systems, has positioned them as vital components in various applications. However, their deployment in IoT devices frequently encounters challenges such as limited hardware capabilities, constrained bandwidth, prolonged data transmission times, and image packet loss due to transmission losses. To address these issues, this paper introduces the Multi-Task learning (MTL) method of Collaborative Intelligence (CI) strategy by dynamically distributing computational tasks among edge devices and cloud. This method addresses the potential performance degradation caused by suboptimal computational splitting points of DNN for multiple tasks (segmentation, classification, depth estimation) and compensates for losses under varying network conditions and data sizes. A key innovation of our methodology is the introduction of a dynamic method to determine split points by computing DNN layers based on real-time bandwidth and data volume. In addition, an Auto Encoder (AE) architecture is implemented in the cloud to reconstruct image data packets lost during transmission based on feature map similarity measurements. Experimental results show that processing all transactions in the cloud with specific operational parameters reduces processing time by 38 % compared to traditional methods, while dynamically selecting the split point results in gains of up to 61 %. Furthermore, the proposed method achieves efficiency by reducing energy consumption by up to 50 % compared to cloud-only processing. It demonstrates robustness under varying network delays and reduces inference time by up to 47.5 % under low-latency conditions. In this regard, the innovative use of an AE for data loss reconstruction also shows significant potential in comp

关键词： Auto encoder Collaborative intelligence deep neural network IoT Multi-task learning Data reconstruction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：