检索结果-内蒙古大学图书馆

Medical Imaging 2023: Biomedical Applications in Molecular, Structural, and Functional Imaging

作者： Hamghalam, Mohammad Do, Richard K. G. Simpson, Amber L. School of Computing Queen's University KingstonON Canada Department of Electrical Engineering Qazvin Branch Islamic Azad University Qazvin Iran Department of Radiology Memorial Sloan Kettering Cancer Center New YorkNY United States Department of Biomedical and Molecular Sciences Queen's University KingstonON Canada

ISBN: (纸本)9781510660410

Small liver lesions common to colorectal liver metastases (CRLMs) are challenging for convolutional neural network (CNN) segmentation models, especially when we have a wide range of slice thicknesses in the computed tomography (CT) scans. Slice thickness of CT images may vary by clinical indication. For example, thinner slices are used for presurgical planning when fine anatomic details of small vessels are required. While keeping the effective radiation dose in patients as low as possible, various slice thicknesses are employed in CRLMs due to their limitations. However, differences in slice thickness across CTs lead to significant performance degradation in CT segmentation models based on CNNs. This paper proposes a novel unsupervised attention-based interpolation model to generate intermediate slices from consecutive triplet slices in CT scans. We integrate segmentation loss during the interpolation model's training to leverage segmentation labels in existing slices to generate middle ones. Unlike common interpolation techniques in CT volumes, our model highlights the regions of interest (liver and lesions) inside the abdominal CT scans in the interpolated slice. Moreover, our model's outputs are consistent with the original input slices while increasing the segmentation performance in two cutting-edge 3D segmentation pipelines. We tested the proposed model on the CRLM dataset to upsample subjects with thick slices and create isotropic volume for our segmentation model. The produced isotropic dataset increases the Dice score in the segmentation of lesions and outperforms other interpolation approaches in terms of interpolation metrics. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Interpolation

来源：评论

学校读者我要写书评

暂无评论

面向嵌入式平台的轻量级光学遥感图像舰船检测

引用

光学学报 2023年第12期43卷 113-126页

作者：王慧赢王春平付强韩子硕张冬冬陆军工程大学石家庄校区电子与光学工程系河北石家庄050003 中国人民解放军32356部队青海西宁710003

针对现有基于深度学习的轻量级目标检测算法对复杂遥感场景图像中舰船目标检测精度低、检测速度慢的问题,提出了一种面向嵌入式平台的轻量级光学遥感图像舰船实时检测算法(STYOLO)。首先,针对主干网络内存访问成本较高的问题,利用高效... 详细信息

针对现有基于深度学习的轻量级目标检测算法对复杂遥感场景图像中舰船目标检测精度低、检测速度慢的问题,提出了一种面向嵌入式平台的轻量级光学遥感图像舰船实时检测算法(STYOLO)。首先,针对主干网络内存访问成本较高的问题,利用高效网络架构ShuffleNet v2作为主干网络对图像进行特征提取,降低内存访问成本,提高网络并行度;其次,利用Slim-neck特征融合结构作为特征增强网络,以融合较低层级特征图中的细节信息,增强对小目标的特征响应,在多尺度信息融合区域施加坐标注意力机制,强化目标关注以提高较难样本检测以及抗背景干扰能力;最后,提出一种跨域迁移和域内迁移相结合的学习策略,减少源域与目标域的差异性,提升迁移学习效果。实验结果表明:基于光学遥感图像舰船检测公开数据集HRSC2016,与同类型快速检测算法YOLOv5s相比,所提算法的检测精度提高了2.7个百分点,参数量减少了61.77%,在嵌入式平台Jetson Nano上检测速度达到102.8 frame/s,能够有效实现对光学遥感图像中舰船目标的实时、准确检测。

关键词：光学遥感图像舰船检测实时检测嵌入式平台注意力机制迁移学习

来源：评论

学校读者我要写书评

暂无评论

A Deep-neural-network-Based Aerosol optical Depth (AOD) Retrieval from Landsat-8 Top of Atmosphere Data

引用

REMOTE SENSING 2022年第6期14卷 1411-1411页

作者： She, Lu Zhang, Hankui K. Bu, Ziqiang Shi, Yun Yang, Lu Zhao, Jintao Ningxia Univ Sch Geog & Planning Yinchuan 750021 Ningxia Peoples R China South Dakota State Univ Geospatial Sci Ctr Excellence Dept Geog & Geospatial Sci Brookings SD 57007 USA Yellow River Conservancy Commiss Yellow River Inst Hydraul Res Zhengzhou 450003 Peoples R China

The 30 m resolution Landsat data have been used for high resolution aerosol optical depth (AOD) retrieval based on radiative transfer models. In this paper, a Landsat-8 AOD retrieval algorithm is proposed based on the deep neural network (DNN). A total of 6390 samples were obtained for model training and validation by collocating 8 years of Landsat-8 top of atmosphere (TOA) data and aerosol robotic network (AERONET) AOD data acquired from 329 AERONET stations over 30 degrees W-160 degrees E and 60 degrees N-60 degrees S. The Google Earth Engine (GEE) cloud-computing platform is used for the collocation to avoid a large download volume of Landsat data. Seventeen predictor variables were used to estimate AOD at 500 nm, including the seven bands TOA reflectance, two bands TOA brightness (BT), solar and viewing zenith and azimuth angles, scattering angle, digital elevation model (DEM), and the meteorological reanalysis total columnar water vapor and ozone concentration. The leave-one-station-out cross-validation showed that the estimated AOD agreed well with AERONET AOD with a correlation coefficient of 0.83, root-mean-square error of 0.15, and approximately 61% AOD retrievals within 0.05 + 20% of the AERONET AOD. Theoretical comparisons with the physical-based methods and the adaptation of the developed DNN method to Sentinel-2 TOA data with a different spectral band configuration are discussed.

关键词： aerosol optical depth (AOD) Landsat-8 deep neural network Google Earth Engine AERONET Collection-2

来源：评论

学校读者我要写书评

暂无评论

Photonics for Sustainable computing

arXiv

引用

arXiv 2024年

作者： Fayza, Farbin Rao, Satyavolu Papa Bunandar, Darius Gupta, Udit Joshi, Ajay Boston University United States NY CREATES United States Lightmatter United States Cornell Tech United States

Photonic integrated circuits are finding use in a variety of applications including optical transceivers, LIDAR, bio-sensing, photonic quantum computing, and Machine Learning (ML). In particular, with the exponentially increasing sizes of ML models, photonics-based accelerators are getting special attention as a sustainable solution because they can perform ML inferences with multiple orders of magnitude higher energy efficiency than CMOS-based accelerators. However, recent studies have shown that hardware manufacturing and infrastructure contribute significantly to the carbon footprint of computing devices, even surpassing the emissions generated during their use. For example, the manufacturing process accounts for 74% of the total carbon emissions from Apple in 2019. This prompts us to ask – if we consider both the embodied (manufacturing) and operational carbon cost of photonics, is it indeed a viable avenue for a sustainable future? So, in this paper, we build a carbon footprint model for photonic chips and investigate the sustainability of photonics-based accelerators by conducting a case study on ADEPT, a photonics-based accelerator for deep neural network inference. Our analysis shows that photonics can reduce both operational and embodied carbon footprints with its high energy efficiency and at least 4× less fabrication carbon cost per unit area than 28 nm CMOS. © 2024, CC BY.

关键词： Carbon footprint

来源：评论

学校读者我要写书评

暂无评论

Violence Detection from CCTV Footage Using optical Flow and Deep Learning in Inconsistent Weather and Lighting Conditions 5th

Violence Detection from CCTV Footage Using Optical Flow and ...

引用

5th International Conference on Advances in computing and Data Sciences (ICACDS)

作者： Madhavan, R. Utkarsh Vidhya, J. V. SRM Inst Sci & Technol Chennai Tamil Nadu India

ISBN: (纸本)9783030814625;9783030814618

Physical assault detection in surveillance systems plays an extremely significant role in the safety of our city. The current methods include a primitive approach for video-based pre-processing methods, which includes grayscale conversion, image augmentation, etc. The models include 3D Convolutional neural networks (3DCNN) and Support Vector Machines (SVMs). The method which we propose consists of a preprocessing method that involves isolating the pixels into a video (or a group of videos) relevant to the action that is taking place in the original video. In addition, a deep learning model which includes a time-distributed Convolutional neural network (CNN) + Long Short Term Memory (LSTM) model extracts the relevant spatial and temporal information or features that can help us predicting violence from an input CCTV footage. Given an input CCTV footage, our model will be able to extract the relevant features to distinguish a violent event from a non-violent one, preferably independent of the lighting. In the current approaches, it was observed that the number of relevant pixels dedicated to the action is low (due to naive cropping). The present system doesn't take into account the inconsistent weather and lighting conditions and its accuracy is expected to drop under those conditions.

关键词： Violence CCTV Physical assault Deep neural network Computer vision CNN LSTM Fight optical flow

来源：评论

学校读者我要写书评

暂无评论

An End-to-End Two-Stream network Based on RGB Flow and Representation Flow for Human Action Recognition

An End-to-End Two-Stream Network Based on RGB Flow and Repre...

引用

Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

作者： Song-Jiang Lai Tsun-Hin Cheung Ka-Chun Fung Tian-Shan Liu Kin-Man Lam Center for Advances in Reliability and Safety New Territories Hong Kong Department of Electrical and Electronic Engineering The Hong Kong Polytechnic University Kowloon Hong Kong

ISBN: (数字)9798350367331

ISBN: (纸本)9798350367348

With the rapid development of deep learning, the performance of computer vision tasks has significantly improved, making the two-stream neural network model a popular research topic for video-based action recognition. Traditional models use RGB and optical-flow streams, which offer promising performance but at high computational cost and complexity. To address these issues, we introduce a representation flow algorithm and replace the optical-flow branch in the egocentric action recognition model with this new branch for end-to-end training. This can greatly reduce computational cost and prediction runtime. Our new model, applied to egocentric action recognition, incorporates class attention maps (CAMs) to enhance recognition accuracy and uses ConvLSTM for spatio-temporal encoding with spatial attention. Evaluated on the GTEA61, EGTEA GAZE+, and HMDB datasets, our model matches the original model’s accuracy on GTEA61 and exceeds it by 0.65% and 0.84% on EGTEA GAZE+ and HMDB, respectively. The proposed model’s runtimes for predicting are significantly faster: 0.1881s, 0.1503s, and 0.1459s compared to the original’s 101.6795s, 25.3799s, and 203.9958s. We also conducted ablation studies to analyze the impact of different parameters on model performance.

关键词： Computer vision Image motion analysis Runtime Accuracy Computational modeling optical computing Predictive models Encoding Computational efficiency Streams

来源：评论

学校读者我要写书评

暂无评论

HarD-CASDE: Segmentation And Lineage Tracking of HeLa Cells

HarD-CASDE: Segmentation And Lineage Tracking of HeLa Cells

引用

International Joint Conference on neural networks (IJCNN)

作者： Fuyun He Shengwen Chen Huiling Feng Guanglian Li School of Electronic and Information Engineering Guangxi Normal University Guilin China Guangxi Key Laboratory of Brain-Inspired Computing and Intelligent Chips Guangxi Normal University Guilin China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

For a long time, cell segmentation and lineage tracing have been one of the focuses of biomedical research. At present, challenges exist in segmentation technology when dealing with cell boundary regions and small fragment cell images, and existing tracking methods underperform in tracking densely populated cell areas. Addressing these issues, we proposes a novel cell segmentation and lineage tracing method, termed HarD-CASDE. This approach merges the benefits of the HarDNet model and cascaded fusion encoder, effectively employing multi-scale information and integrating contextual features. It employs four types of harmonic densely connected networks (HarDBlock) as encoders and introduces a multi-resolution cascaded fusion encoder module. It improves boundary detection and feature segmentation, thereby enhancing the overall performance of the segmentation task. Furthermore, to enhance the segmentation of small targets, we use a combined loss function based on wBCE and wIOU to train the network. Experiments on the HeLa cell dataset show that our HarD-CASDE model significantly improves segmentation performance compared to models like Unet and TransUnet. Compared to four other classic tracking methods, our approach achieves a tracking accuracy of 0.9176. Moreover, the study verifies the significance of each module in cell segmentation and tracking tasks through ablation experiments.

关键词： Image segmentation Solid modeling Accuracy Target tracking Three-dimensional displays neural networks optical computing

来源：评论

学校读者我要写书评

暂无评论

Optic Disc Segmentation in Fundus Images Using U-Net

Optic Disc Segmentation in Fundus Images Using U-Net

引用

International Conference on New Trends in computing Sciences (ICTCS)

作者： Nour Alhendi Hadi Hamad Mathematics Department An-Najah National University Nablus Palestine

ISBN: (数字)9798331523411

ISBN: (纸本)9798331523428

Optic disc (OD) is one of the most important parts of the retina. Accurate segmentation of the OD provides essential information about the health of the retina and aids in the diagnosis of various diseases that affect the retina, such as glaucoma and diabetic retinopathy (DR). Segmenting the OD manually can be influenced by numerous variables like image clarity and lighting, demanding expertise for precise boundary detection. The nature of this task emphasizes the need for accessible automated or semi-automated OD segmentation solutions to address the shortage of specialists in diagnosing retinal conditions effectively almost worldwide. In this study, an automatic method for OD segmentation in retinal images using a convolutional neural network (CNN) architecture, known as U-Net, was introduced. To enhance the U-Net's efficacy compared to previous studies, we adopted a two-step strategy: first, by cropping the image around the OD using the bounding box technique, and then by resizing it to $128 \times 128$ pixels. This method not only preserved robust OD segmentation but also significantly reduced computational time remarkably. Then, these images were enhanced using the contrast limited adaptive histogram equalization (CLAHE) to eliminate the noise and improve their qualities. After that, a U-Net model was constructed and trained to obtain segmented images. The proposed model was trained and evaluated using the public dataset ORIGA, and the predicted results were compared with the ground truth (GT) images. This method competed with other studies and achieved competing results, average value of each of the following metrics came to be: recall 95.33%, F-score 96.54%, and IoU 93.41%.

关键词： Integrated optics Training Image segmentation Adaptation models Accuracy Computational modeling Computer architecture Retina optical imaging Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Hausa Handwriting Character Recognition Using CNN and Tesseract 2

Hausa Handwriting Character Recognition Using CNN and Tesser...

引用

2nd International Conference on Advanced Innovations in Smart Cities, ICAISC 2025

作者： Dauda, Muhammad Khamis Musa, Umar Al-Barhamtoshy, Hassanin Faculty of Computing and Information Technology Department of Information Technology Jeddah Saudi Arabia

ISBN: (纸本)9798331506995

The Hausa language, spoken by over 64 million people, is a vital part of West Africa's cultural and social fabric. Despite its importance, handwritten Hausa script recognition has little concern due to its tonal nature, unique characters, and limited datasets. This study tackles these challenges by creating a dataset and evaluating two models commonly used in handwritten recognition: a Convolutional neural network (CNN) and Tesseract optical Character Recognition (OCR). The custom CNN, trained on data from 30 volunteers with varied writing styles, achieved an accuracy of 96%, with strong precision, recall, and F1-score metrics. In contrast, Tesseract OCR struggled, particularly with Hausa-specific characters like , , , and , achieving lowest accuracy for these classes. Although it had non-zero recognition in some characters, its performance was inconsistent, indicating difficulties in generalization. This work provides a scalable solution for handwritten Hausa character recognition, contributing to research on low-resource African languages and paving the way for better access to Hausa handwritten literature and integration into natural language processing applications. © 2025 IEEE.

关键词： optical character recognition

来源：评论

学校读者我要写书评

暂无评论

SAGNet: Synergistic Attention-Graph network For video salient object detection

引用

Image and Vision computing 2025年 160卷

作者： Lina, Huo Gao, Xueyuan Wang, Wei Chen, Ke Wang, Ke College of Computer and Cyber Security Hebei Normal University Shijiazhuang050024 China

In the field of video salient object detection (VSOD), accurately capturing motion information is essential. Previous approaches primarily rely on optical flow, convolutional long short term memory (ConvLSTM), or 3D convolutional neural network (CNN) to extract and utilize motion information. However, these methods capture limited motion details and increase the parameters in the network. Moreover, Transformer-based methods, while effective in high-level feature modeling, suffer from excessive computational complexity and insufficient local feature extraction, limiting their practical application in VSOD. To address these challenges, we propose a novel synergistic attention-graph network (SAGNet) that independently distills spatial–temporal cues and spatial edge features using the synergistic attention-graph module (SAGM) and the spatial edge attention module (SEM), respectively. SAGM innovatively integrates inter-frame attention with spatial–temporal graph convolution network (GCN). The inter-frame attention proposed in SAGM captures motion information between video frames while expanding the receptive field to capture long-range dependencies. Spatial–temporal GCN models video as a graph, bridge features from temporal into spatial branch, which is capable of fusing cross-modal features collaboratively. This synergy enables SAGNet to consider both global and local spatial–temporal features. SEM enhances high-level information by extracting spatial and edge features from the low-level data using the Sobel operator and spatial attention module. Experimental results on several publicly available VSOD benchmark datasets demonstrate that SAGNet outperforms existing methods in terms of detection accuracy and efficiency, confirming its superiority and practicality in VSOD. © 2025

关键词： Object detection

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：