检索结果-内蒙古大学图书馆

16th International Symposium on Applied Reconfigurable Computing

作者： Ali, Muhammad Rad, Pedram Amini Goehringer, Diana Tech Univ Dresden Dresden Germany

ISBN: (纸本)9783030445331;9783030445348

Modern image processing applications, like object detection or image segmentation, require high computation and have high memory requirements. For ASIC-/FPGA-based architectures, hardware accelerators are a promising solution, but they lack flexibility and programmability. To fulfill flexibility, computational and memory intensive characteristics of these applications in embedded systems, we propose a modular and flexible RISC-v based MPSoC architecture on Xilinx Zynq Ultrascale+ MPSoC. The proposed architecture can be ported to other Xilinx FPGAs. Two neural networks (Lenet-5 and Cifar-10 example) were used as test applications to evaluate the proposed MPSoC architectures. To increase the performance and efficiency, different optimization techniques were adapted on the MPSoC and results were evaluated. 16-bit fixed-point parameters were used to have a compression of 50% in data size and algorithms were parallelized and mapped on the proposed MPSoC to achieve higher performance. A 4x parallelization of a NN algorithm on the proposed MPSoC resulted in 3.96x speed up and consumed 3.61x less energy as compared to a single soft-core processor setup.

关键词： MPSoC NoC RISC-v FPGA SoC Power estimation

来源：评论

学校读者我要写书评

暂无评论

Implementation of a Multitudinous Face Recognition using YOLO.v3

Implementation of a Multitudinous Face Recognition using YOL...

引用

Microelectronics, Signals & systems (ICMSS), International Conference on

作者： M Suman Menon Anju George N Aswathy Adi Shankara Institute of Engineering and Technology Kalady India

ISBN: (纸本)9781665448864

Face recognition is one of the most functional research in present scenario, with many practical and commercial applications including identification, access control, forensics, medical care, human-computer interactions, security, etc. Face recognition technique is rapidly becoming the mainstay of state of the art technological security solution. One of the crucial applications of face recognition in the current scenario is linked with security. Identifying people from a crowd or a group of people require an exceptional algorithm. One of the most arduous tasks about the existing face recognition system is the processing or prediction time. The current systems focus on accuracy than speed, which leads to an increase in the detection time. There are several techniques in machine learning and deep learning. But deep learning is preferred more than machine learning for detection and recognition applications because of the large availability of data. An algorithm for fast real-time object detecting and recognizing application is required. YOLO (you only look once) is a single shot deep learning object detection algorithm. In this work, the working of the YOLO algorithm and implementing multiple face recognition using YOLO version 3 is explained. A custom dataset is created from taken from Kaggle and google. At the time of testing the model, a processing speed of 30 ms was obtained.

关键词： Deep learning Machine learning algorithms image recognition Pandemics Face recognition Object detection Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

An active contour model for medical image segmentation using a quaternion framework 18

An active contour model for medical image segmentation using...

引用

18th image processing: algorithms and systems Conference, IPAS 2020

作者： voronin, v. Zhdanova, M. Semenishchev, E. Zelensky, A. Agaian, S. Moscow State University of Technology "STANKIN" Moscow Russia Don State Technical University Lab. «Mathematical methods of image processing and computer vision Intelligent Systems» Gagarina 1 Rostov on Don Russia CUNY/ College of Staten Island Dept. of Computer Science New York United States

This paper presents a new method for segmenting medical images is based on Hamiltonian quaternions and the associative algebra, method of the active contour model and LPA-ICI (local polynomial approximation - the intersection of confidence intervals) anisotropic gradient. Since for segmentation tasks, the image is usually converted to grayscale, this leads to the loss of important information about color, saturation, and other important information associated color. To solve this problem, we use the quaternion framework to represent a color image to consider all three channels simultaneously when segmenting the RGB image. As a method of noise reduction, adaptive filtering based on local polynomial estimates using the ICI rule is used. The presented new approach allows obtaining clearer and more detailed boundaries of objects of interest. The experiments performed on real medical images (Z-line detection) show that our segmentation method of more efficient compared with the current state-of-art methods. © 2020, Society for Imaging Science and Technology

关键词： Adaptive filtering

来源：评论

学校读者我要写书评

暂无评论

Machine learning technology for neoplasm segmentation on brain mri scans

Machine learning technology for neoplasm segmentation on bra...

引用

2020 PhD Symposium at ICT in Education, Research, and Industrial Applications, ICTERI-PhD 2020

作者： Sova, Ivan Sidenko, Ievgen Kondratenko, Yuriy Intelligent Information Systems Department Petro Mohyla Black Sea National University 68th Desantnykiv Str. 10 Mykolaiv54003 Ukraine

In this paper, machine learning technology for neoplasm segmentation on brain MRI scans is analyzed. This analysis allows to choose the most appropriate machine learning architecture and various preprocessing techniques to increase the precision of tumor instance segmentation. Understanding the image and extracting information from it to accomplish some result is an important area of application in digital image technology. image segmentation has quickly found its use in medicine and specifically oncology. Precise segmentation masks may not be critical in other cases, but marginal segmentation errors in medical images may render the results unreliable for clinical use. Therefore, biomedical problems require much higher boundary detection precision to improve further analysis. Comparison of different machine learning algorithms, neural network architectures will achieve the highest accuracy of recognition and segmentation. During the comparison, a system of the U-Net architecture with additional processing methods was selected as the final model. Its accuracy reached 94%, which is a significant result compared to manual image segmentation. Copyright © 2020 for this paper by its authors.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

image processing system in conditions of uncertainty and the training of its operator

Image processing system in conditions of uncertainty and the...

引用

International Workshop on Advanced Technologies in Material Science, Mechanical and Automation Engineering (MIP) - Engineering

作者： Alekseev, v. v. Lakomov, D. v. Maamari, A. G. Shishkin, A. A. Petrukhnova, G. v. Tambov State Tech Univ 106 Sovetskaya St Tambov 392000 Russia Voronezh State Tech Univ 14 Moscow Ave Voronezh Russia

In modern information systems, decision making based on image processing is hampered by the impact of negative external and internal factors leading to image blurring, which introduces uncertainty in this process. In this regard, algorithms and models are used to reduce the effect of uncertainty in image analysis. The article presents a new adaptive algorithm for image processing in different wave bands. The article also presents the results of research on the training of operators of image processing systems in conditions of uncertainty. It is proposed to train the operators of these systems on the basis of a competence-based approach using an information system that allows you to create individual training paths for the operators. The implementation of the training information system is proposed to be made on the basis of a web service.

关键词： Information systems

来源：评论

学校读者我要写书评

暂无评论

Corner detection of intensity images with cellular neural networks (CNN) and evolutionary techniques

引用

NEUROCOMPUTING 2019年 347卷 82-93页

作者： Cuevas, Erik Diaz-Cortes, Margarita-Arimatea Mezura-Montes, Efren Univ Guadalajara Dept Elect CUCEI Av Revoluc 1500 Guadalajara 44430 Jalisco Mexico FU Berlin Inst Informat Arnimallee 7 D-14195 Berlin Germany Univ Veracruz Artificial Intelligence Res Ctr Sebastian Camacho 5 Xalapa 91090 Veracruz Mexico

Corner detection represents one of the most important steps to identify features in images. Due to their powerful local processing capabilities, Cellular Nonlinear/Neural Networks (CNN) are commonly utilized in image processing applications such as image edge detection, image encoding and image hole filling. CNN perform well for locating corner features in binary images. However, their use in grayscale images has not been considered due to their design difficulties. In this paper, a corner detector based on CNN for grayscale images is presented. In the approach, the original processing scheme of the CNN is modified to include a nonlinear operation for increasing the contrast of the local information in the image. With this adaptation, the final CNN parameters that allow the appropriate detection of corner points are estimated through the Differential evolution algorithm by using standard training images. Different test images have been used to evaluate the performance of the proposed corner detector. Its results are also compared with popular corner methods from the literature. Computational simulations demonstrate that the proposed CNN approach presents competitive results in comparison with other algorithms in terms of accuracy and robustness. (C) 2019 Elsevier B.v. All rights reserved.

关键词： Cellular neural/nonlinear networks Corner detection Differential evolution Nonlinear systems image processing

来源：评论

学校读者我要写书评

暂无评论

Recognition of Abnormal Red Blood Cells Using Earth Mover's Distance Algorithm

Recognition of Abnormal Red Blood Cells Using Earth Mover's ...

引用

International Conference on ent Computing and Control systems (ICICCS)

作者： Mikaela-Kashmir B. Lazaro Y. Alfredo Lumot John Maynard M. Marquita v. Manuelito Misa Bryan Angelo A. Reotutar Roselito E. Tolentino Polytechnic University of the Philippines-Santa Rosa Santa Rosa Philippines

Red blood cells (RBCs) play a significant role by carrying essential nutrients and oxygen throughout the body. These cells are shaped as biconcave discs that are flexible enough to pass through small capillaries easily. However, if these cells are irregularly shaped, they may reduce the oxygen-carrying capacity of the blood and have difficulty moving through the blood vessels. Blood tests are laboratory methods that are often performed by medical experts to manually determine abnormalities in the blood, but it may be subjective and susceptible to human error. Therefore, recognizing abnormal RBCs automatically through digital image processing would be beneficial. There are existing studies that focus on the application of different image processing techniques in recognizing abnormal RBCs automatically. However, they suffer from certain weaknesses such as: others did not cover all the abnormal RBCs, while the others still encountered a significant error in recognition due to the slim dissimilarities between the attributes that they used. In this study, the proponents sought to improve and provide solution for the limitations of the previous studies by proposing a system that can recognize 9 abnormal RBCs using a machine learning algorithm called Earth Mover's Distance Algorithm which allows partial matching between two different distributions. The test images used in the system are microscopic images of blood samples gathered from the web. The images gathered initially undergone image preprocessing to extract the necessary attributes from multiple regions of interests (ROIs) needed in the computation of the Earth Mover's Distance (EMD). EMD is the measure of dissimilarity between two multi-dimensional distributions in space. The lowest computed EMD represents the smallest dissimilarity of the input image to the template image. As a result, the system displays the total number and label of each abnormal RBCs recognized in a Graphic User Interface (GUI). The proposed sy

关键词： Earth image recognition Machine learning algorithms Red blood cells Microscopy Process control Extraterrestrial measurements

来源：评论

学校读者我要写书评

暂无评论

Detection of Hotspots in Fluorescence Imaging of Yeast Cell Model used in Neuro-Degenerative Research

Detection of Hotspots in Fluorescence Imaging of Yeast Cell ...

引用

Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT), International Conference on

作者： S N Chandrasekar Abhishek Rao G Sai Muthukumar v venketesh S R Raghunatha Sarma Sri Sathya Sai Institute of Higher Learning (SSSIHL) Prashanthi Nilayam Andhra Pradesh INDIA

ISBN: (数字)9781728157917

ISBN: (纸本)9781728157924

Fluorescence (FL) optical imaging modality is an indispensable tool in biomedical research for the discovery of drugs as well as diagnosis. With recent developments in high throughput microscopy hardware platforms, there is an immense need to match and develop automated image processing tools that can aid clinical pathologists and researchers expedite their workflow and analysis. This plays a crucial role when the volume of samples is immense and the workflow is tedious manually. In this report, we successfully demonstrate the ability of our software tool to very specifically detect hotspots in FL images of yeast-based model systems that are employed specifically for the detection of neurodegenerative disorders in systems biology research. The algorithm demonstrated here can achieve more feature extraction than many of the reported morphometric tools available till date particularly while using Fluorescence modality.

关键词： image segmentation Microscopy Biological system modeling Software algorithms Fluorescence Tools Throughput

来源：评论

学校读者我要写书评

暂无评论

NeuTomPy toolbox, a Python package for tomographic data processing and reconstruction

引用

SOFTWAREX 2019年 9卷 260-264页

作者： Micieli, Davide Minniti, Triestino Gorini, Giuseppe Univ Calabria Dipartimento Fis Via Pietro Bucci I-87036 Cosenza Italy Univ Milano Bicocca Dipartimento Fis G Occhialini Piazza Sci 3 I-20126 Milan Italy Rutherford Appleton Lab STFC ISIS Facil Harwell OX11 0QX Berks England

In this article we present the NeuTomPy Toolbox, a new Python package for tomographic data processing and reconstruction. The toolbox includes pre-processing algorithms, artifacts removal and a wide range of iterative reconstruction methods as well as the Filtered Back Projection algorithm. The NeuTomPy toolbox was conceived primarily for neutron tomography datasets and developed to support the need of users and researchers to compare state-of-the-art reconstruction methods and choose the optimal data processing workflow for their data. In fact, in several cases sparse-view datasets are acquired to reduce scan time during a neutron tomography experiment. Hence, there is great interest in improving quality of the reconstructed images by means of iterative methods and advanced image-processing algorithms. The toolbox has a modular design, multi-threading capabilities and it supports Windows, Linux and Mac OS operating systems. The NeuTomPy toolbox is open source and it is released under the GNU General Public License v3, encouraging researchers and developers to contribute. In this paper we present an overview of the main toolbox functionalities and finally we show a typical usage example. (C) 2019 The Authors. Published by Elsevier B.v.

关键词： Tomographic reconstruction software Tomography Neutron imaging

来源：评论

学校读者我要写书评

暂无评论

GPU-Accelerated Adaptive Dictionary Learning and Sparse Representations for Multispectral image Super-resolution

GPU-Accelerated Adaptive Dictionary Learning and Sparse Repr...

引用

IEEE India Conference (INDICON)

作者： Trishna Barman Bhabesh Deka A.v.v. Prasad Tezpur University Tezpur India Natational Remote Sensing Center (NRSC) ISRO Hyderabad Telengana India

ISBN: (纸本)9781665441766

Recently, single image super-resolution (SISR) based on sparse representations has been gaining much attention from the research community in the field of remote sensing. In this paper, a fast SISR reconstruction framework is developed for multispectral remote sensing (MSRS) images based on adaptive dictionary learning and sparse representations. It consists of two major parts: first, a novel super-resolution approach is developed for MSRS using sparse coding and adaptive dictionary learning. High-frequency features present in the input low-resolution MS image are extracted by using Butterworth low-pass, difference of Gaussian (DoG), and Sobel filters in horizontal and vertical directions. The proposed feature extraction method reveals the edges and other detailed information present in the MS image effectively. Secondly, massively parallel algorithms are designed for adaptive dictionary learning and sparse reconstruction using the Compute Unified Device Architecture (CUDA)-enabled General Purpose-Graphics processing Unit (GP-GPU) programming model. The proposed method GP-GPU implementation not only gives better results in terms of visual quality and objective fidelity criteria, but also significantly reduces the computation time compared to its CPU counterparts to achieve near-real time operating speed.

关键词： visualization image edge detection Superresolution Machine learning Programming Feature extraction Real-time systems

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：