检索结果-内蒙古大学图书馆

6th International Conference on Information Systems and Computer Networks, ISCON 2023

作者： Khandelwal, Ankit Shukla, Aasheesh Sain, Mangal GLA University Electronics and Communication Dept. Mathura India Dongseo University Information Engineering Dept. Busan Korea Republic of

ISBN: (纸本)9798350346961

This paper presents a review on automated disease detection processing. The primary issue in herbal plants is the diagnosis and stratification of its diseases. The conventional process is unpredictable and inconsistent, therefore techniques for classifying plants diseases using digital method are necessary as they speed up the identification process. This literature discusses various digital methods for classifying plant diseases. The first step is image processing of herbal plants, which make use of traditional image processing techniques and algorithms to categories diseases in herbal plants based on attributes like texture, color, and shape. The second step is plant categorization based on deep learning and machine learning, such as CNN, KNN, logistic regression etc. Deep learning applications mainly lessen the reliance on designs and preprocessing techniques by joining the entire process. This review provides a basic understanding and knowledge of the current state-of-the-art researches, as well as some dynamic trends for future work.. © 2023 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

FeSViBS: Federated Split Learning of vision Transformer with Block Sampling 26th

FeSViBS: Federated Split Learning of Vision Transformer with...

引用

26th International Conference on Medical image Computing and Computer-Assisted Intervention (MICCAI)

作者： Almalik, Faris Alkhunaizi, Naif Almakky, Ibrahim Nandakumar, Karthik Mohamed Bin Zayed Univ Artificial Intelligence Abu Dhabi U Arab Emirates

ISBN: (纸本)9783031438943;9783031438950

Data scarcity is a significant obstacle hindering the learning of powerful machine learning models in critical healthcare applications. Data-sharing mechanisms among multiple entities (e.g., hospitals) can accelerate model training and yield more accurate predictions. Recently, approaches such as Federated Learning (FL) and Split Learning (SL) have facilitated collaboration without the need to exchange private data. In this work, we propose a framework for medical imaging classification tasks called Federated Split learning of vision transformer with Block Sampling (FeSViBS). The FeSViBS framework builds upon the existing federated split vision transformer and introduces a block sampling module, which leverages intermediate features extracted by the vision Transformer (ViT) at the server. This is achieved by sampling features (patch tokens) from an intermediate transformer block and distilling their information content into a pseudo class token before passing them back to the client. These pseudo class tokens serve as an effective feature augmentation strategy and enhances the generalizability of the learned model. We demonstrate the utility of our proposed method compared to other SL and FL approaches on three publicly available medical imaging datasets: HAM1000, BloodMNIST, and Fed-ISIC2019, under both iiD and non-iiD settings. Code: https://***/faresmalik/FeSViBS.

关键词： Split learning Federated learning vision transformer Convolutional neural network Augmentation Sampling

来源：评论

学校读者我要写书评

暂无评论

In-Scene Material Detection for Real-Time Autonomous Hyperspectral Reflectance Correction 11

In-Scene Material Detection for Real-Time Autonomous Hypersp...

引用

Conference on image Sensing Technologies - Materials, Devices, Systems, and applications XI

作者： Pietersen, Randall A. Robinson, Brian M. Diltz, Robert A. Einstein, Herbert H. MIT Dept Civil & Environm Engn Room 1-342 77 Massachusetts Ave Cambridge MA 02139 USA USAF Civil Engn Ctr 104 Res Rd Bldg T9719 Tyndall AFB FL 32403 USA Torch Technol 4090 South Mem Pkwy Huntsville AL 35802 USA

ISBN: (纸本)9781510673793;9781510673786

If a U.S. Air Force operated airfield is attacked, the current methodology for assessing its condition is a slow manual inspection process, exposing personnel to dangerous conditions. Advances in drone technology, remote sensing, deep learning, and computer vision have sparked interest in developing autonomous remote solutions. While digital image processing techniques have matured in recent decades, a lack of application-specific training data presents significant obstacles for developing reliable solutions to detect specific objects amongst rubble, debris, variations in pavement types, changing surface features, and other variable runway conditions. Consequently, near-surface hyperspectral imaging has been proposed as an alternative to RGB digital images, due to its discriminatory power in classifying materials. Spatio-spectral data acquired by hyperspectral imagers help address common challenges presented by data scarcity and scene complexity;however, raw data acquired by hyperspectral sensors must first undergo a reflectance correction process before it can be of use. This paper presents an expedient method, tailored to airfield damage assessment, for performing autonomous reflectance correction on near-surface hyperspectral data using in-scene pavement materials with a known spectral reflectance. Unlike most reflectance correction methods, this process eliminates the need for human intervention with the sensor (or its data) pre or post flight and does not require pre-staged reference targets or an additional downwelling irradiance sensor. Positive initial results from real-world flights over pavements are presented and compared to traditional methods of reflectance correction. Three separate flight tests report mean errors between 2% and 2.5% using the new method.

关键词： Hyperspectral Imaging Pavement Reflectance Correction machine Learning Computer vision

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale ConvLSTM Attention-Based Brain Tumor Segmentation

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND applications 2022年第11期13卷 849-856页

作者： Ait Skourt, Brahim Nikolov, Nikola S. Majda, Aicha Begdouri, Ahlame Univ Sidi Mohammed Ben Abdellah Lab Intelligent Syst & Applicat Fes Morocco Univ Limerick Comp Sci & Informat Syst Dept Limerick Ireland Univ Moulay Ismail Networks & Comp Syst Res Team Meknes Morocco

In computer vision, there are various machine learning algorithms that have proven to be very effective. Convolutional Neural Networks (CNNs) are a kind of deep learning algorithms that became mostly used in image processing with a remarkable success rate compared to conventional machine learning algorithms. CNNs are widely used in different computer vision fields, especially in the medical domain. In this study, we perform a semantic brain tumor segmentation using a novel deep learning architecture we called multi-scale ConvLSTM Attention Neural Network, that resides in Convolutional Long-Short-Term-Memory (ConvLSTM) and Attention units with the use of multiple feature extraction blocks such as Inception, Squeeze-Excitation and Residual Network block. The use of such blocks separately is known to boost the performance of the model, in our case we show that their combination has also a beneficial effect on the accuracy. Experimental results show that our model performs brain tumor segmentation effectively compared to standard U-Net, Attention U-net and Fully Connected Network (FCN), with 79.78 Dice score using our method compared to 78.61, 73.65 and 72.89 using Attention U-net, standard U-net and FCN respectively.

关键词： Convolutional neural networks image processing semantic brain tumor segmentation convolutional long short term memory inception squeeze-excitation residual-network attention units

来源：评论

学校读者我要写书评

暂无评论

A machine vision Approach for Bioreactor Foam Sensing

引用

SLAS TECHNOLOGY 2021年第4期26卷 408-414页

作者： Austerjost, Jonas Soeldner, Robert Edlund, Christoffer Trygg, Johan Pollard, David Sjoegren, Rickard Sartorius Stedim Biotech GmbH Sartorius Corp Res August Spindler Str 11 D-37079 Gottingen Germany Sartorius Stedim Data Analyt AB Sartorius Corp Res Umea Sweden Sartorius Stedim North Amer Inc Sartorius Corp Res Boston MA USA

machine vision is a powerful technology that has become increasingly popular and accurate during the last decade due to rapid advances in the field of machine learning. The majority of machine vision applications are currently found in consumer electronics, automotive applications, and quality control, yet the potential for bioprocessing applications is tremendous. For instance, detecting and controlling foam emergence is important for all upstream bioprocesses, but the lack of robust foam sensing often leads to batch failures from foam-outs or overaddition of antifoam agents. Here, we report a new low-cost, flexible, and reliable foam sensor concept for bioreactor applications. The concept applies convolutional neural networks (CNNs), a state-of-the-art machine learning system for image processing. The implemented method shows high accuracy for both binary foam detection (foam/no foam) and fine-grained classification of foam levels.

关键词： foam sensor machine vision deep learning process analytical technology bioprocessing

来源：评论

学校读者我要写书评

暂无评论

Gaze is more than just a point: Rethinking visual attention analysis using peripheral vision-based gaze mapping 23

Gaze is more than just a point: Rethinking visual attention ...

引用

15th Annual ACM Symposium on Eye Tracking Research and applications (ETRA)

作者： Wang, Felix Sihan Lohmeyer, Quentin Duchowski, Andrew Meboldt, Mirko Swiss Fed Inst Technol Inst Design Mat & Fabricat Zurich Switzerland Clemson Univ Sch Comp Clemson SC USA

ISBN: (纸本)9798400701504

In mobile eye-tracking, visual attention is commonly evaluated using fixation-based measures, which can be mapped to predefined objects of interest for task-specific attention analysis. Even though attention can be directed independently from the fovea, little research can be found on the quantification of peripheral vision for attention analysis. In this work, we discuss the benefits of enhancing traditional mapping methods with near-peripheral information and expand previous research by presenting a novel machine learning-based gaze measure, the visual attention index (VAI), for the analysis of visual attention using dynamic stimuli. Results are discussed using the data of two multi-object mobile eye tracking use cases and visualized using radar graphs. We show that by combining foveal and peripheral vision the VAI is effective for the comparison of visual attention over multiple tasks, trials and subjects, which offers new possibilities for a more realistic and detailed depiction of visual attention in multi-object tasks.

关键词： Attention visualization peripheral vision Computer vision image processing Eye movements and cognition AOI-based methods Visual attention eye movement measures

来源：评论

学校读者我要写书评

暂无评论

Using Monocular Depth Estimation for Distance Estimation in a Moving Vehicle 1

Using Monocular Depth Estimation for Distance Estimation in ...

引用

1st International Conference on Computer Graphics and image processing, CGIP 2023

作者： De Guzman, Lanz Benedict N. Raymond See, Aaron De la Salle University Manufacturing Engineering and Management Department Manila City Philippines Southern Taiwan University of Science and Technology Department of Electrical Engineering Tainan City Taiwan

ISBN: (纸本)9798350333183

Accompanying the increase in demand for autonomous systems and robotic solutions is the increase in the relevance of various depth estimation technologies. Monocular Depth Estimation (MDE) is used to predict distances by generating depth maps using only a single RGB camera. However, without out-of-the-box calibration or ground truth reference for generated depth values from MDE models its use case in practical applications is limited. This research introduces a method of actualizing generated depth map values for different applications. The proposed system involves the utilization of machine vision using YOLO for object detection, followed by the computation of the lens optic algorithms to calculate the distance. Results demonstrated a real-time environment detection and depth estimation solution with more than 90% accuracy for measuring object depth in static environments. Furthermore, the system was also successfully tested in a moving vehicle to provide an estimated distance of surrounding vehicles. In the future, further tests will be done to improve the accuracy and calculation speed for use in car safety. © 2023 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Computing Resistance-Style image Sensors for Artificial Neural Networks

引用

IEEE INTERNET OF THINGS JOURNAL 2023年第6期10卷 4985-4997页

作者： Zhao, Guihua Peng, Yating Wang, Yizhan Wan, Caihua Liu, Xianping Zhang, Yu Han, Xiufeng Chen, Weichong Yu, Zhiyi Sun Yat Sen Univ Sch Elect & Informat Technol Guangzhou 510006 Peoples R China Sun Yat Sen Univ Sch Microelect Sci & Technol Zhuhai 519082 Peoples R China Chinese Acad Sci Univ Chinese Acad Sci Inst Phys Beijing Natl Lab Condensed Matter Phys Beijing 100190 Peoples R China Songshan Lake Mat Lab Magnet Lab Dongguan 523808 Guangdong Peoples R China Univ Chinese Acad Sci Ctr Mat Sci & Optoelect Engn Beijing 100049 Peoples R China Sun Yat Sen Univ Guangdong Prov Key Lab Optoelect Informat Proc Chi Guangzhou 510275 Peoples R China

Today, machine vision experiences large latency due to big data processing, which is a barrier to time-critical applications. To address this issue, in-sensor computing was presented in the past. Here, we present a scheme of computing in a magnetic tunneling junction (MTJ) sensor array for proof-of-principle. Using the MTJ sensor array, the functions of artificial neural network (ANN) classifiers and autoencoders were verified. The time for correct classification of one picture was less than 9 mu s . The power consumed in the sensor array can be decreased according to the square law without affecting the results. Our work shows universal circuits and algorithms to compute in resistance-style ANN image sensors with promising energy efficiency.

关键词： Manganese Magnetic tunneling Magnetic resonance imaging Sensor arrays Saturation magnetization Magnetic circuits Magnetic domains Artificial neural network (ANN) image sensor in-sensor computing magnetic tunneling junction (MTJ) sensor design sensor system integration

来源：评论

学校读者我要写书评

暂无评论

vision-Based Mobile Robots Control Along a Given Trajectory 22nd

Vision-Based Mobile Robots Control Along a Given Trajectory

引用

22nd International Conference on Artificial Intelligence and Soft Computing (ICAISC)

作者： Rodziewicz-Bielewicz, Jan Korzen, Marcin West Pomeranian Univ Technol Szczecin Ul Zolnierska 49 PL-71210 Szczecin Poland

ISBN: (纸本)9783031425073;9783031425080

The paper presents the application of a computer vision approach to tracking the mobile robot's state. As an exemplary environment, we use a feedback control system for the trajectory planning and control. The system in the feedback loop use images taken from a centrally placed camera and, based on this, calculates the robots states, i.e. position and angle of rotation. The solution is adopted for indoor experiments. The experimental part shows the application of trajectory planning for multiple robots to cover a given area. The robot state is calculated using the YOLO model. We show that current machine learning techniques are fast and accurate for such applications and do not require image preprocessing or camera calibration.

关键词： Mobile robotics Computer vision vision-based control Visual servoing iRobot YOLO Trajectory planing

来源：评论

学校读者我要写书评

暂无评论

Visual Feedback System Supporting Robotic Manipulation of Hemp Plants

引用

JOURNAL OF NATURAL FIBERS 2025年第1期22卷

作者： Kraft, Marek Ptak, Bartosz Piechocki, Mateusz Pieczynski, Dominik Mlodzikowski, Kamil Kulecki, Bartlomiej Belter, Dominik Poznan Univ Tech Inst Robot & Machine Intelligence Piotrowo 3A PL-60965 Poznan Poland

This paper presents an agricultural robotics system designed to automate the detection and manipulation of male hemp plants, addressing the challenge of manually removing these to enhance crop quality. The solution uses an industry-standard, inexpensive RGB-D camera as the input data source to derive control signals controlling the robotic arm and end effector. Input image data processing is performed by a dedicated neural network model trained using a dataset created specifically for the described task to achieve detection by stalk segmentation and postprocessing. The research involved assessing various neural network models, including UNet, DeepLabV3+, and YOLOv8 in various variants, for their capability to detect stalks accurately and swiftly. Fast operation is necessary for effective real-time feedback in robotic grasping tasks. Among tested architectures, the integration of UNet with ResNet50 was found to provide a good trade-off between detection precision and operational speed on edge AI devices. The resulting solution offers good accuracy and significantly outperforms existing methods in terms of processing speed, promising substantial improvements in agricultural robotics by enabling on-line adaptive grasping using low-cost components. The applications can be extended beyond hemp tending to include various other crops, eliminating tedious manual labor. (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic). (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic),(sic)(sic)(sic)RGB-D(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic). (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(si

关键词： Hemp computer vision deep learning robotics manipulation (sic)(sic) (sic)(sic)(sic)(sic)(sic) (sic)(sic)(sic)(sic) (sic)(sic)(sic) (sic)(sic)

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：