检索结果-内蒙古大学图书馆

Development of methods for parallel processing of series of images obtained by a machine vision system in various electromagnetic ranges 2

Development of methods for parallel processing of series of ...

引用

Conference on Optical Instrument Science, Technology, and applications II Held as Part of SPIE Optical System Design Conference

作者： Semenishchev, Evgenii Voronin, Viacheslav Alepko, Andrey Zelensky, Aleksandr Agaian, Sos Moscow State Tech Univ STANKIN Moscow Russia CUNY Dept Comp Sci New York NY 10314 USA

ISBN: (数字)9781510645974

ISBN: (纸本)9781510645974;9781510645967

machine vision systems used in modern industrial complexes, based on the analysis of multi and hyperspectral imaging. The transition to implementing the "Industry 4.0" program is not possible when using one type of data. The first control system used only the visible range image. They made it possible to analyze the trajectories of movement of objects, control product quality, carry out security functions (control of perimeter crossing), etc. The development of new industrial robotic cells and processing complexes using cognitive functions implying the receipt, analysis, and processing of heterogeneous data. The construction of a unified information field, which allows performing multidimensional operations with data, allows increasing the speed of decision-making and the implementation of automated robot-human systems at the level of an assistant working in a unified workspace. The use of machine vision systems analyzing information received in: visible (shape, the trajectory of movement, position of objects, etc.);near-infrared range (data is similar to visible, allows operation in dusty, foggy, low light conditions);far-infrared range - thermal (plotting temperature gradients, identifying areas of overheating);ultraviolet range (analysis of ionization sources, corona discharges, static charges, tags);X-ray and microwave ranges (analysis of the surface and internal structure of objects, allow the identification of defects);range and 3D sensors (construction of volumetric figures, analysis of the relative position of objects and their interaction), etc. Data analysis is often performed not by a single camera but by a group of sensors located not in a single housing. Primary data integration reduces the number of information channels while maintaining the functionality and accuracy of the analysis. The article discusses creating fusion images obtained by industrial sensors into a combined image containing joint data. Combining multi and hyperspectral imaging makes i

关键词： preprocessing IR image local features denoising combining image fusion image

来源：评论

学校读者我要写书评

暂无评论

EG-SNIK: A Free Viewing Egocentric Gaze Dataset and Its applications

引用

IEEE ACCESS 2022年 10卷 129626-129641页

作者： Malladi, Sai Phani Kumar Mukherjee, Jayanta Larabi, Mohamed-Chaker Chaudhury, Santanu Indian Inst Technol Kharagpur Adv Technol Dev Ctr Kharagpur 721302 India IIT Kharagpur Dept Comp Sci & Engn Kharagpur 721302 India Univ Poitiers Xlim Umr Cnrs 7252 F-86000 Poitiers France IIT Jodhpur Dept Comp Sci & Engn Jodhpur 342030 India

Egocentric vision data captures the first person perspective of a visual stimulus and helps study the gaze behavior in more natural contexts. In this work, we propose a new dataset collected in a free viewing style with an end-to-end data processing pipeline. A group of 25 participants provided their gaze information wearing Tobii Pro Glasses 2 set up at a museum. The gaze stream is post-processed for handling missing or incoherent information. The corresponding video stream is clipped into 20 videos corresponding to 20 museum exhibits and compensated for user's unwanted head movements. Based on the velocity of directional shifts of the eye, the I-VT algorithm classifies the eye movements into either fixations or saccades. Representative scanpaths are built by generalizing multiple viewers' gazing styles for all exhibits. Therefore, it is a dataset with both the individual gazing styles of many viewers and the generic trend followed by all of them towards a museum exhibit. The application of our dataset is demonstrated for characterizing the inherent gaze dynamics using state trajectory estimator based on ancestor sampling (STEAS) model in solving gaze data classification and retrieval problems. This dataset can also be used for addressing problems like segmentation, summarization using both conventional machine and deep learning approaches.

关键词： Egocentric gaze dataset museum exhibit head movement compensation representative scanpath categorization

来源：评论

学校读者我要写书评

暂无评论

SmartSeg: machine learning infrastructure software for accelerating medical image segmentation in patient-specific applications

引用

COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING-IMAGING AND VISUALIZATION 2024年第1期12卷

作者： Burton, William Gibbons, Kalin Economon, Keaghan Seg AI LLC Denver CO 80229 USA Boise State Univ Mech & Biomed Engn Comp Biosci Lab Boise ID USA

Patient-specific applications in biomechanics and orthopaedics call for segmentation of volumetric medical images. This task has historically been performed manually or semi-automatically, which entails significant effort by trained experts. Computer vision algorithms based on machine learning have emerged as an effective tool for accelerating medical image processing tasks through automation. The current work describes fully no-code software which supports development of custom machine learning models for automatic segmentation of volumetric medical images. This cloud-integrated desktop application allows researchers and biomedical engineers to define segmentation algorithms, configure annotated datasets, and train personalised machine learning algorithms with cloud compute in just a few clicks. A presented case study demonstrates an ideal workflow in the software using femoral cartilage segmentation from magnetic resonance imaging as a representative use case. Models developed in the software demonstrated mean dice similarity coefficients of up to 0.886 on a test cohort, which is competitive with previously reported methods developed with significantly larger data sets. The mean surface difference between digitised models reconstructed from ground truth and predicted segmentation was 0.33 mm. Results suggest the described software enables creation of accurate machine learning models with limited engineering effort.

关键词： machine learning automatic segmentation medical images

来源：评论

学校读者我要写书评

暂无评论

An automatic machine vision-based algorithm for inspection of hardwood flooring defects during manufacturing

引用

ENGINEERING applications OF ARTIFICIAL INTELLIGENCE 2023年第PartA期123卷

作者： Truong, Van Doi Xia, Jiaping Jeong, Yuhyeong Yoon, Jonghun Hanyang Univ Dept Mech Design Engn Seongdonggu 222 Wangsimni Ro Seoul 04763 South Korea Hanyang Univ Dept Mech Engn 55 Hanyangdaehak Ro Ansan 15588 Gyeonggi Do South Korea Hanyang Univ BK21 FOUR ERICA ACE Ctr Ansan 15588 Gyeonggi South Korea

Hardwood flooring products are popular construction materials because of their aesthetics, durability, low maintenance requirements, and affordability. To ensure product quality during manufacturing, common defects such as cracks, chips, or stains are typically detected and classified manually, but this process can decrease productivity. The aim of this study was to develop an automatic machine vision-based inspection system with a robust algorithm for inspecting small hardwood flooring defects in a production line. This defect-inspection algorithm is based on image-processing techniques, including background elimination, boundary approximation, and defect inspection of photographs. The YOLOv5 deep-learning algorithm for object detection was applied to detect surface defects. The resulting algorithm identified the quality of each specimen (i.e., either good or defective). The influences of colour and surface patterns on defect inspection were experimentally investigated under light conditions. The algorithm was adaptable to specimens with different colours and patterns under various conditions, demonstrating the potential of this approach in practical situations.

关键词： Hardwood flooring Automatic defect inspection image processing Yolov5

来源：评论

学校读者我要写书评

暂无评论

A Survey on Audio-Video Based Defect Detection Through Deep Learning in Railway Maintenance

引用

IEEE ACCESS 2022年 10卷 65376-65400页

作者： DE DONATO, L. O. R. E. N. Z. O. FLAMMINI, F. R. A. N. C. E. S. C. O. MARRONE, S. T. E. F. A. N. O. MAZZARIELLO, C. L. A. U. D. I. O. NARDONE, R. O. B. E. R. T. O. SANSONE, C. A. R. L. O. VITTORINI, V. A. L. E. R. I. A. Univ Naples Federico II Dept Elect Engn & Informat Technol I-80125 Naples Italy Linnaeus Univ Dept Comp Sci & Media Technol S-35195 Vaxjo Sweden Hitachi Rail STS Digital & Data Driven Innovat Unit I-80147 Naples Italy Univ Naples Parthenope Dept Engn I-80143 Naples Italy

Within Artificial Intelligence, Deep Learning (DL) represents a paradigm that has been showing unprecedented performance in image and audio processing by supporting or even replacing humans in defect and anomaly detection. The railway sector is expected to benefit from DL applications, especially in predictive maintenance applications, where smart audio and video sensors can be leveraged yet kept distinct from safety-critical functions. Such separation is crucial, as it allows for improving system dependability with no impact on its safety certification. This is further supported by the development of DL in other transportation domains, such as automotive and avionics, opening for knowledge transfer opportunities and highlighting the potential of such a paradigm in railways. In order to summarize the recent state-of-the-art while inquiring about future opportunities, this paper reviews DL approaches for the analysis of data generated by acoustic and visual sensors in railway maintenance applications that have been published until August 31st, 2021. In this paper, the current state of the research is investigated and evaluated using a structured and systematic method, in order to highlight promising approaches and successful applications, as well as to identify available datasets, current limitations, open issues, challenges, and recommendations about future research directions.

关键词： Rail transportation Maintenance engineering Rails Sensors Inspection Task analysis Focusing Computer vision machine learning fault detection inspection CNN smart railways

来源：评论

学校读者我要写书评

暂无评论

Hydrogel-Scalable Nanoslide for Switchable Optical Spatial-Frequency processing

引用

LASER & PHOTONICS REVIEWS 2023年第4期17卷 2200368-2200368页

作者： Dai, Chenjie Li, Zhe Sh, Yangyang Wan, Shuai Hu, Wanlin Li, Zhongyang Wuhan Univ Elect Informat Sch Wuhan 430072 Peoples R China Wuhan Inst Quantum Technol Wuhan 430206 Peoples R China Wuhan Univ Sch Microelect Wuhan 430072 Peoples R China Wuhan Univ Suzhou Inst Suzhou 215123 Peoples R China

Spatial frequency processing is an essential technology for extracting morphological information from an optical image. Although various Fourier-based flat optical elements have been proposed as spatial-frequency filters to realize image processing, their transfer functions are statically fixed once fabricated, thus limiting the versatile, dynamic functionalities and practical applications. Here, a novel practical tuning strategy is demonstrated to realize switchable spatial-frequency processing for edge-enhanced and bright-field imaging by employing a tunable hydrogel-scalable nanoslide. By utilizing multilayered metallic and hydrogel stacks construction, the nanoslide directly manipulates the optical spatial frequency in the wavevector domain and exhibits opposite image processing at different wavelength channels due to the cavity-induced wavelength-sensitivity. More intriguingly, via controlling the ambient humidity, the angular-dependent optical response of the nanoslide can be effectively tuned for dynamic edge-enhanced imaging due to the hydrogel's inflation from moisture. In addition, the nanoslide is readily fabricated at a large scale and integrated into compact imaging systems, such as a biomicroscope. With the advantages of a high numerical aperture approximate to 0.8, polarization-insensitive, microscopy-compatible, and facile architecture, the proposed hydrogel-based nanoslide can find potential applications in machine vision, real-time image processing, biological imaging, and analog computing.

关键词： humidity-responsive hydrogel-scalable nanoslides spatial-frequency processing switchable wavevector manipulation

来源：评论

学校读者我要写书评

暂无评论

Enhancing Algorithmic Robustness through Noise-Induced Dataset Augmentation for Fusion Fuel Profile Inversion

Enhancing Algorithmic Robustness through Noise-Induced Datas...

引用

image processing, Computer vision and machine Learning (ICICML), International Conference on

作者： Ke Xu Jun Zhao Guoliang Xiao XingZhong Xiong Artificial Intelligence Key Laboratory of Sichuan Province Sichuan University of Science and Engineering Yinbin China Southwestern Institute of Physics Chengdu China

ISBN: (数字)9798350355413

ISBN: (纸本)9798350355420

In the field of nuclear fusion fueling, the distribution of fuel beam parameters is of great significance for optimizing injection techniques and for in-depth study of the interaction between particles and plasma. To obtain the beam velocity distribution of the supersonic molecular fueling beam, a U-net model has been applied in our previous research, successfully capturing the mapping relationship between density distribution and velocity distribution, and effectively predicts velocity distributions across a wide range of parameters. To effectively apply this model to the inversion of experimental data, it is imperative to rigorously investigate its noise resistance capabilities. This paper focuses on enhancing algorithmic robustness through noise-induced dataset augmentation. The results suggest that under 20% noise interference, the model can still achieve stable prediction of the velocity distribution of the supersonic molecular beam, providing strong support for subsequent research.

关键词： Resistance Particle beams Gaussian noise Molecular beam applications Predictive models Fusion reactors Feature extraction Prediction algorithms Robustness Fuels

来源：评论

学校读者我要写书评

暂无评论

Research on Evaluation Method of Aerial image Segmentation Algorithm 7

Research on Evaluation Method of Aerial Image Segmentation A...

引用

7th International Conference on Signal and image processing, ICSIP 2022

作者： Wang, Heng Zhuang, Chuangang Zhao, Jiamei Shi, Rui Jiang, Hai Yuan, Yanrong Guo, Xinyi Xue, Zhichao China Academy of Launch Vehicle Technology Science and Technology on Space Physics Laboratory Beijing100076 China

ISBN: (数字)9781665495639

ISBN: (纸本)9781665495639

Aerial image segmentation algorithms are used in many scenarios such as traffic warning, road assistance, terrain mapping and military reconnaissance. It has been widely applied in both military and civilian applications. A lot of researches on target detection and recognition technology in aerial images have been carried out. In addition, technologies such as computer vision and machine learning have been introduced. Compared with the research on image segmentation algorithms, there are relatively few studies on evaluation method segmentation algorithms. In view of the above situation, the evaluation method of aerial image segmentation is established, and the performance of comprehensive evaluation of the medhod for image segmentation algorithms is verified. Specifically, it includes similarity index, misidentification similarity index and distance index. Selecting image data sources with different degrees of complexity are selected in order to verify the effectiveness of the above image segmentation evaluation method. The results show that the evaluation indicators prove that the algorithms can achieve effective target segmentation to a certain extent. The segmentation accuracy rate is above 95%. In view of the accuracy and effectiveness of the aerial image segmentation algorithm, it is expected to be further applied in the field of aerial images. © 2022 IEEE.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

From Templates to Transformers: A survey of Multimodal image Captioning Decoders

From Templates to Transformers: A survey of Multimodal Image...

引用

Computer, Electronics & Electrical Engineering & their applications (IC2E3), International Conference on

作者： Himanshu Sharma Devanand Padha Department of Computer Science and IT Central University of Jammu Rahya-Suchani Jammu India

Automated text-to-image (image synthesis) and image-to-text (image captioning) generation are two of the most challenging and cutting-edge fields of study in Computer vision (CV) in conjunction with Natural Language processing (NLP). The image-to-text synthesis, also known as image Captioning (IC), has numerous applications in visual assistance, machine vision, remote security, healthcare, and remote sensing, among others. Since their origin, the IC frameworks have been constructed utilizing a two-subcomponent pipeline consisting of the visual feature extraction and natural language modeling subcomponents. The recent surge of interest in the applications of sequential deep learning models in machine translation has resulted in the development of efficient language modeling architectures. Deep sequential modeling architectures, such as Recurrent Neural Networks (RNN), Long Short Term Memory (LSTM), and Gated Recurrent Units (GRU), tackle the intricacies of the multi-modular space far better than existing machine learning-based translation approaches. The tremendous growth of deep language modeling architectures in IC necessitates a comprehensive review of its literature. In this survey, we conduct an exhaustive and analytical analysis of the language modeling architectures used in IC frameworks for caption generation, along with their training corpus datasets. We also identify a list of open research issues and potential research areas for future work.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generating Caption From images Using Flickr image Dataset

Generating Caption From Images Using Flickr Image Dataset

引用

International Conference on Computing and Networking Technology (ICCNT)

作者： Piyush Sharma Kiranpreet Kaur Aaryan Gandotra Pranay Patial Radhika Sehaj Gaba Department of CSE Chandigarh University Mohali India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

In an age where the Internet is dominated by visual content, the generation of animated captions has become a must. It has always been an interesting study for researchers in the Department of Artificial Intelligence. Enabling the machine to describe images with the same skillful accuracy as the human has important applications in various fields such as robotic vision, manufacturing, and beyond This project integrates recurrent neural networks with is a topic of contextual parallelism dedicated to extracting features from images Natural Language processing Computer vision and integrating them seamlessly, this research provides insights a it goes further on this interdisciplinary topic. Additionally, annotations for the sample images are created and performed a comparative analysis of different feature extraction and encoder patterns to determine which model provided the highest accuracy and delivered the desired results.

关键词： Visualization Accuracy Recurrent neural networks image databases Computational modeling Parallel processing Feature extraction Web sites Multimedia communication Object recognition

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：