检索结果-内蒙古大学图书馆

2025 International conference on Advancement in Data Science, E-learning and Information System, ICADEIS 2025

作者： Syahda, Tabina Shafa Nabila Ma'ady, Mochamad Nizar Palefi Lidiawaty, Berlian Rahmy Telkom University Surabaya Department of Information Systems Surabaya Indonesia

ISBN: (纸本)9798331513320

Artificial Intelligence (AI) in ophthalmology has been growing, driven by the increasing volume of clinical data that can be utilized for algorithm development. image processing on fundus disease is crucial in providing visual information for cataract detection. Therefore, the application of technology that can leverage fundus images for cataract detection is still a few. In this regard, several previous studies have explored CNN algorithms for automatic cataract screening using fundus images. However, other research groups have proposed or developed new algorithms as alternatives. This research compares two CNN models, VGG16 and MobileNet, while also incorporating additional approaches such as Ensemble Learning and Hybrid Method. VGG16 and MobileNet were chosen because they are known for their high accuracy performance and computational efficiency. Source data obtained from Kaggle as many as 1000 datasets. The results of the study indicate that the model with the Ensemble Learning approach achieved the highest average accuracy, followed by MobileNet, MobileNet + SVM, VGG16, and VGG16 + SVM, with accuracies of 0.9377;0.9150;0.9143;0.9129;and 0.9104, respectively. This research is expected to aid medical professionals in diagnosing cataracts more accurately and guide engineers in selecting the best machine-learning models for fundus image-based cataract detection systems. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Dark area recognition algorithm for SAR images based on superpixel statistical features

Dark area recognition algorithm for SAR images based on supe...

引用

2024 Advanced Fiber Laser conference, AFL 2024

作者： Liu, Peng Liu, Yanyang Li, Xuemao Dong, Xuezhi xiao, Qian Li, Zhenfang National Key Laboratory of Radar Signal Processing Xidian University Xi'an710071 China Shanghai Institute of Satellite Engineering Shanghai200090 China Aerospace Stellar Technology Co. Ltd. Beijing100095 China

ISBN: (数字)9781510688889

ISBN: (纸本)9781510688872

The dark areas in SAR images are primarily low backscatter regions dominated by noise signals, characterized by significant texture, low echo energy, and large phase interference errors in the echo signals. Accurately identifying these dark regions is crucial for high-precision SAR image target detection and classification, SAR cross-radiometric calibration of satellite constellations, SAR image filtering, and the generation of digital surface models based on SAR technology. This paper proposes a dark region recognition algorithm for SAR images based on superpixel statistical features(SSF). The algorithm constructs a bimodal distribution curve using the statistical features of superpixels and searches for the region with the minimum amplitude proportion within the bimodal distribution, which serves as a constraint to ensure that the OTSU thresholding method converges to the global optimal solution. This approach addresses the issue of inaccurate threshold detection in traditional algorithms, significantly improving the accuracy of dark region detection in SAR images. The detection results can further be used to identify low backscatter areas, such as shadows, water bodies, and asphalt roads. © 2025 SPIE.

关键词： Superpixels

来源：评论

学校读者我要写书评

暂无评论

CIGraph: Accelerating Graph Queries over Database with Compressed Index 24th

CIGraph: Accelerating Graph Queries over Database with Com...

引用

24th International conference on algorithms and Architectures for Parallel processing, ICA3PP 2024

作者： Lv, Zhen Wang, Li Liu, Yingfan Zhao, Kankan Peng, Yanguo Fundamentals Department Air Force Engineering University Xi’an China School of Computer Science and Technology Xidian University Xi’an China Shanghai Yunxi Technology Co. Ltd. Shanghai China

ISBN: (纸本)9789819615476

Graph database management systems (GDBMS) can efficiently store and deal with massive graph data that reflect complex relations among entities. However, most existing works can not adapt well to the query requirements of large-scale complex graph data. In this paper, a prototype CIGraph is presented based on a compressed index, which preserves a high compression ratio and query performance on graph data. Through sufficient experimental evaluations on LDBC-SNB benchmark, CIGraph with compressed index outperforms JanusGraph in aspects of both compression ratio and query performance. Compared with JanusGraph, CIGraph occupies less than a third of the space to store the same graph data, and the acceleration coefficient for graph algorithms is up to 3X on average in CIGraph. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Graph Databases

来源：评论

学校读者我要写书评

暂无评论

A multi-dimension information processing method for low-power forest surveillance systems 3

A multi-dimension information processing method for low-powe...

引用

2024 3rd International conference on Aerospace and Control Engineering, ICoACE 2024

作者： Wan, Zijing Tianjin Fire Science and Technology Research Institute MEM Weijinnan Road No. 110 Tianjin Nankai300170 China

To identify early-stage forest fires using ground-based sensing systems, this paper studies the radiation characteristics of smoke particles and interference particles under natural light in different environments based on a rapid dimensional reduction algorithm of forest monitoring images and a multi-dimensional time series signal graph model. It has completed the forest fire classification and spread prediction method based on multi-sensor and image sequence signal abnormal event detection, which enables the forest fire detection system to achieve comprehensive and efficient collection of fire information at a lower cost and indirectly reduces sensor power consumption and bandwidth. The results can provide quantifiable solutions for forest fire detection and disaster situation assessment and are suitable for the comprehensive promotion of on-site sensing systems. © 2025 Institute of Physics Publishing. All rights reserved.

关键词： Fire detectors

来源：评论

学校读者我要写书评

暂无评论

Research on Radar Imaging Method Based on Tensor Neural Network 16

Research on Radar Imaging Method Based on Tensor Neural Netw...

引用

16th International conference on Signal processing systems, ICSPS 2024

作者： Qiu, Tian Liu, Nan Yang, Minglei National Key Laboratory of Radar Signal Processing Xidian University Xian710071 China

ISBN: (纸本)9781510689251

LFM signals, as the most commonly used form of signals in radar systems, are of significant research value. At this stage, compressed sensing (CS) has significant potential to improve the image quality of radar imaging. However, the existing CS theories have limited ability to improve the imaging capability of LFM signals, and always suffer from low imaging efficiency, parameters need to be manually adjusted, and poor suppression of side flaps. In this paper, we propose a neural network imaging algorithm based on tensor compressed perception, called 2D-ISTA-Net. this algorithm accomplishes distance-azimuth 2D radar imaging by combining a model-based iterative threshold shrinkage (ISTA) algorithm and a data-driven deep learning model. First, the sparse recovery model of the LFM signal is obtained analytically, and then, the 2D-ISTA optimization algorithm is unfolded into a deep network structure in which all parameters are optimized based on data-driven auto-learning instead of manual tuning. Subsequently, the network learning is trained on the scattering point images in the training set to obtain a 2D-ISTA-Net network with autonomous parameter tuning capability. Furthermore, we compared this algorithm with some existing CS algorithms. The data show that 2D-ISTA-Net significantly improves the computing power and has a lower side flap level compared with the traditional CS algorithm. Experiments show that 2D-ISTA-Net has good reconstruction performance. © 2025 SPIE.

关键词： Radar imaging

来源：评论

学校读者我要写书评

暂无评论

Synergistic Integration of MRI Prior for Enhanced Photoacoustic Tomography: Advancing Biomedical Imaging Quality 12th

Synergistic Integration of MRI Prior for Enhanced Photoacous...

引用

12th International conference on Frontiers of Intelligent Computing: Theory and Applications, FICTA 2024

作者： Krishna Satya Varma, M. Yamini Priyanka, S. Department of IT SRKR Engineering College Chinaamiram Andhra Pradesh Bhimavaram India

ISBN: (纸本)9789819601424

In this research endeavor, we introduce a pioneering strategy to enhance the fidelity of photoacoustic tomography (PAT) images, addressing prevalent artifacts and distortions stemming from acoustic and optical properties of biological tissues. Our approach capitalizes on the synergy between PAT and magnetic resonance imaging (MRI), exploiting MRI's wealth of information on tissue anatomy and composition to refine PAT reconstructions. Through a meticulously designed framework, we integrate advanced image processing methodologies, spanning registration, fusion, and deep learning-based algorithms, to seamlessly merge MRI priors into the PAT reconstruction pipeline. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： image fusion

来源：评论

学校读者我要写书评

暂无评论

Advanced HLS for Reconfigurable image processing in Big Data systems 12th

Advanced HLS for Reconfigurable Image Processing in Big Da...

引用

12th International conference on Big Data and Artificial Intelligence, BDA 2024

作者： Bhagirath, K. Pant, Dheeraj Tiwari, Abhishek Khaneja, Vivek Noida201309 India

ISBN: (纸本)9783031818202

This paper introduces novel HLS techniques for reconfigurable and memory-efficient image processing within deep learning frameworks, addressing inherent limitations of current deep learning accelerators (DLAs) due to their dependency on the host processor, leading to pipeline delays and reduced throughput. The proposed approach emphasizes parallelization, on-chip buffering, and scalability, enhancing computational efficiency and throughput for AI applications in big data and edge computing environments. Methodologically, the paper emphasizes mathematical analysis, hardware architecture specification, HLS modeling, Verilog RTL implementation. By addressing these challenges, this work contributes to advancing the performance capabilities of DLAs in deep learning inference tasks. The algorithms integrates seamlessly with DLA IP subsystems, such as Axi interfaces, external (DRAM) and internal (SRAM) memory interfaces, and activation/pooling engines. Key contributions include advanced algorithm design to enhance memory efficiency and reconfigurability, essential for adapting to diverse hardware architectures. Experimental validation demonstrates substantial improvements in computational speed and resource utilization compared to traditional methods, making it suitable for intensive AI tasks in big data analytics. In conclusion, this research presents a framework for integrating HLS techniques into hardware accelerators, facilitating efficient image processing across diverse applications in big data and edge computing. These advancements aim to bolster DLAs’ capabilities in handling complex AI workloads effectively, promising enhanced solutions for AI-driven domains. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Post-processing of Thresholding or Deep Learning Methods for Enhanced Tissue Segmentation of Whole-Slide Histopathological images

Post-Processing of Thresholding or Deep Learning Methods for...

引用

18th International Joint conference on Biomedical Engineering systems and Technologies, BIOSTEC 2025

作者： Marczyk, Michal Wrobel, Agata Merta, Julia Polanska, Joanna Department of Data Science and Engineering Silesian University of Technology Gliwice44-100 Poland Yale Cancer Center Yale School of Medicine New HavenCT06511 United States Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice44-100 Poland

Digital pathology allows for the efficient storage and advanced computational analysis of stained histopathological slides of various tissues. Tissue segmentation is a crucial first step of digital pathology aimed at eliminating background, pen markings, and other artifacts, reducing image size, and increasing the efficiency of further analysis. In most cases, color thresholding or deep learning models are used, but their effectiveness is reduced due to complex artifacts and huge color variations between slides. We propose a post processing method to increase the tissue segmentation performance of any initial segmentation algorithm. Using a set of 197 manually annotated histopathological images of breast cancer patients and 63 images of endometrial cancer patients, we tested our method with 3 thresholding techniques and 3 deep learning-based algorithms by calculating the Dice index, Jaccard index, precision, and recall. In both datasets, applying post processing increased precision and recall for thresholding methods and mostly precision for deep learning models. Overall, applying post-processing gave better tissue segmentation performance than initial segmentation methods, significantly increasing Dice and Jaccard indices. Our results proved that thanks to post-processing, the tissue segmentation pipeline is more robust to noises and artifacts commonly present in histopathological images. © 2025 by SCITEPRESS– Science and Technology Publications, Lda.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Fractal image Compression Using Discrete Cosine Transform with Domain Classification and Hexagonal Partitioning

Fractal Image Compression Using Discrete Cosine Transform wi...

引用

International conference on Power Engineering and Intelligent systems, PEIS 2024

作者： Kulkarni, Milind Bharadwaj, Rakhi Joshi Hedaoo, Anushka Vishwakarma Institute of Technology Pune India

ISBN: (纸本)9789819767090

image compression has seen much advancement over the years, with new algorithms and techniques being developed to make image files smaller and easier to transmit and store using both lossy compression methods as well as lossless compression techniques. This paper proposes a lossy fractal image compression method using hexagonal partitioning and domain classification on the quantized image pixels to alleviate the image compression technique's processing complexity. Fractal image compression is a technique for compressing digital images that uses the idea of self-similarity, which is the concept that a pattern can repeat itself at different scales. Our proposed method consists of transforming the spatial domain to frequency domain using discrete cosine transform with quantization and then applying hexagonal partitioning on the square image pixels. A range block is compared with every other classified domain block and the domain block having the least root mean squared (RMS) error is mapped to the range block and replaced with it, storing the original contrast and brightness settings. In this manner, an encoding file is created which stores information of all the mappings, and then the original image is decoded iteratively until the output resembles the original image. To evaluate the performance of the suggested algorithm, numerous simulations are performed. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Discrete cosine transforms

来源：评论

学校读者我要写书评

暂无评论

Adaptive HSV enhancement algorithm for high grayscale X-ray films

引用

Jisuanji Jicheng Zhizao xitong/Computer Integrated Manufacturing systems, CIMS 2025年第4期31卷 1237-1246页

作者： Li, Liangliang Wang, Peng Ren, Jia Lyu, Zhigang Luo, Cheng Li, xiaoyan School of Mechatronic Engineering Xi'an Technological University Xi'an710021 China School of Electronics and Information Engineering Xi'an Technological University Xi'an710021 China School of Information and Communication Engineering Hainan University Haikou570228 China

Aiming at the existing high grayscale image enhancement algorithms with poor robustness and insufficient adaptive capability, an enhancement algorithm capable of adaptively enhancing the processed high-bit images was proposed. To compensate for the brightness and darkness of the original film, an adaptive correction algorithm based on the grayscale prior was designed, and a new high-level HSV pseudo-color enhancement algorithm was presented. In addition, to ensure the balance of the enhanced image, an adaptive balance processing algorithm based on the brightness Channel was proposed. Finally, the enhancement effect was performed on the 12-bit X-ray negative image generated in the industrial welding manufacturing field. The experimental results showed that the proposed method could achieve the best enhancement effect on many different test samples without significantly increasing the complexity of the algorithm. © 2025 CIMS. All rights reserved.

关键词： Photointerpretation

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：