检索结果-内蒙古大学图书馆

4th IEEE Eurasia conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2022

作者： Solomons, Cassandra D. Mathew, Ammu Anna Karthikeyan, V. Shanmugasundaram, Vivekanandan Thomas, Hannah Mary Sathya, A. Rose, Jeyanth School of Electrical Engineering Vellore Institute of Technology Vellore India Christian Medical College Dept. of Radiation Oncology Vellore India

ISBN: (数字)9781728195797

ISBN: (纸本)9781728195797

Anterior Segment OCT is a valuable tool in providing B scan images of the cornea and anterior segment of the eye. Corneal scarring is a residue of various ocular diseases which include corneal ulcers, trauma and dystrophies. Corneal scarring results in corneal blindness and severe visual disability for the affected patient. Current grades of assessing corneal scarring are purely qualitative in nature. Though these clinical grades (nebula, macula, leucoma) are useful in guiding clinicians in making clinical decisions regarding treatment, they lack precision because of the absence of numeric data. Experimental research involving corneal scarring would greatly benefit from quantitative methods which are amenable to statistical analysis. Purpose: To determine the quantum of corneal scarring by image analysis of Anterior Segment Optical Coherence Tomographic (AS-OCT) vertical and horizontal scans of both the scarred and normal corneas of 32 subjects. methods: image processing and pixel analysis of the vertical and horizontal AS-OCT scans were done in the normal and scarred eyes to determine the mean pixel intensity, corneal thickness and quantum of corneal scarring. Results: The mean pixel intensity for scarred corneas was determined to be 126.2 for horizontal and 136.17 for vertical and 100.09 for horizontal, 96.75 for vertical on the normal corneas. The mean corneal thickness on scarred eyes was 204.46 and 217.38 for horizontal and vertical. Whereas on the normal corneas was 202.78 and 201.09 for horizontal and vertical. Quantum of variability for the scarred eyes was 26.11 and 39.42 for horizontal and vertical, respectively. The differences in medians of quantums of three scarred groups were statistically significant. © 2022 IEEE.

关键词： image analysis

来源：评论

学校读者我要写书评

暂无评论

Research on the Prediction Model of University Library Borrowing Behavior Based on Big Data and Machine Learning

Research on the Prediction Model of University Library Borro...

引用

image processing and Computer Applications (ICIPCA), IEEE International conference on

作者： Xin Zhang Desheng Tian Hubei University of Technology Wuhan China

ISBN: (数字)9798350360240

ISBN: (纸本)9798350384161

The use of machine learning algorithms for deep data mining is aimed at improving the efficiency of public library borrowing recommendations. According to the characteristics of users' reading behavior, a heuristic recommendation method which is more suitable for users' use behavior is designed. Using RBF network to predict the information of book borrowing, a method to predict the number of book borrowing is constructed. The method is compared with other prediction methods. The results show that this method can estimate the quantity of collection accurately and reduce the error rate of estimation. The prediction method proposed in this paper is effective. The recommendation algorithm model shows high application efficiency in simulation test and trial run test, and the subjective evaluation results are given by members who have used the new system. Through the analysis of literature retrieval results, empirical evidence confirms the viability of the search process paradigm predicated on heuristic discovery presented within this document.

关键词： Technological innovation Accuracy statistical analysis Estimation Predictive models Big Data Libraries Data models Data mining Resource management

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Approach for Detection of Diabetic Retinopathy with Improved Pre-processing

Machine Learning Approach for Detection of Diabetic Retinopa...

引用

International conference on Computing, Communication, and Intelligent Systems (ICCCIS)

作者： Sharma, Ayushi Shinde, Swapnil Shaikh, Imran Ismail Vyas, Madhav Rani, Soumya Ramrao Adik Inst Technol Dept Informat Technol Nerul India

ISBN: (纸本)9781728185293

Diabetic Retinopathy has been one of the common reason why diabetic patients lose sight. It is caused by the damaging of the blood vessel of the light sensitive tissue at the back of the eye or the retina. The traditional methods for the detection of the disease requires lots of time and efforts hence the proposed system aims to detect diabetic retinopathy using image processing and Machine Learning. The input provided to the proposed system are the retinal images from the standard databases DIARETDB0 and DIARETDB1. The system aims to automate the process of detection using a combination of basic image processing steps with more focus on pre-processing to obtain clear image for feature extraction, further Machine Learning algorithms are applied for classification. The statistical features such as area and perimeter are obtained from blood vessels and exudates after pre-processing stage, different pre-processing techniques such as grayscale conversion, binarization, canny edge detection and some morphological operations including image dilation and erosion are applied to obtain clear image. The Machine Learning algorithms that are used for result analysis are Weighted KNN, Cubic SVM and Simple Tree which provides an accuracy of 85.8%, 87.2% and 88.6% respectively. The sensitivity and specificity parameters are also obtained, Simple tree yields better results for all the three result parameters.

关键词： Blood Vessels Diabetic Retinopathy Exudates Machine Learning Weighted KNN

来源：评论

学校读者我要写书评

暂无评论

Digital Halftone Classification using Simplified CNN and stochastic Statistics

Digital Halftone Classification using Simplified CNN and Sto...

引用

Asia-Pacific-Signal-and-Information-processing-Association Annual Summit and conference (APSIPA ASC)

作者： Guo, Jing-Ming Sankarasrinivasan, S. Natl Taiwan Univ Sci & Technol Taipei Taiwan

ISBN: (纸本)9789881476890

Digital halftone technique is widely utilized in various printing applications to generate high quality printing output. The halftone classification is a fundamental requirement for the optimal reconstruction of binary printed images during inverse halftoning and image authentication. In this study, a new classification scheme based on convolutional neural networks (CCN) is proposed. For effective training, a new approach exploiting the properties of stochastic halftone parameters is proposed to obtain the favorable halftone patch. The CNN network has been optimized to perform classification with minimal layers, and the model is trained using cross entropy, and make use of softmax classification layer. During testing, the sub-halftone patches are classified individually and then majority voting strategy is performed to identify the actual halftone class. For performance evaluation, a deep learning halftone database has been developed for the 24 halftone varieties which also includes the latest multitone images. From experimental result, it has been validated that the proposed architecture achieves average correct classification rate (ACCR) of 100% for several halftone class, and showed superior performance in comparison with state-of-the-art methods.

关键词： CNN classifier digital half-toning stochastic statistics printing

来源：评论

学校读者我要写书评

暂无评论

statistical investigations into the geometry and homology of random programs

arXiv

引用

arXiv 2024年

作者： Sporring, Jon Larsen, Ken Friis University of Copenhagen Department of Computer Science Copenhagen Denmark

AI-supported programming has taken giant leaps with tools such as Meta’s Llama and openAI’s chatGPT. These are examples of stochastic sources of programs and have already greatly influenced how we produce code and teach programming. If we consider input to such models as a stochastic source, a natural question is, what is the relation between the input and the output distributions, between the chatGPT prompt and the resulting program? In this paper, we will show how the relation between random Python programs generated from chatGPT can be described geometrically and topologically using Tree-edit distances between the program’s syntax trees and without explicit modeling of the underlying space. A popular approach to studying high-dimensional samples in a metric space is to use low-dimensional embedding using, e.g., multidimensional scaling. Such methods imply errors depending on the data and dimension of the embedding space. In this article, we propose to restrict such projection methods to purely visualization purposes and instead use geometric summary statistics, methods from spatial point statistics, and topological data analysis to characterize the configurations of random programs that do not rely on embedding approximations. To demonstrate their usefulness, we compare two publicly available models: ChatGPT-4 and TinyLlama, on a simple problem related to image processing. Application areas include understanding how questions should be asked to obtain useful programs;measuring how consistently a given large language model answers;and comparing the different large language models as a programming assistant. Finally, we speculate that our approach may in the future give new insights into the structure of programming languages. Copyright © 2024, The Authors. All rights reserved.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

An Optimized Ant Gardient Convolutional Neural Network for Disease Detection in Apple Leaves

An Optimized Ant Gardient Convolutional Neural Network for D...

引用

Innovation in Technology (INOCON), IEEE International conference for

作者： Arshleen Kaur Raman Chadha Computer Science and Engineering Chandigarh University Gharuan Mohali (Punjab) India

This article implements a new method of classification or detection of the diseased section in apple leaf images. Apple leaf disease segmentation and classification is the main processing phase for ROI extraction in agriculture-based IP (image processing). Generally, in the case of live images, it is also crucial to separate the apple leaf images from the background. A frequent experiment in processing leaf images is the automated separation of apple leaves from the background. In the research work, ALS (Alternaria leaf spot), BS (brown spot), Mosaic, etc., are five normal categories of ALD (apple leaf diseases) that rigorously affect apple production. Though the previous work lacks accuracy and faster detection of apple diseases to confirm the healthy growth of apple production, it implements a DL method based on the CNN classifier for detecting apple leaf diseases. The OAGCNN detection model is proposed, which combines an ant colony optimizer and stochastic gradient descent method with the CNN model. The simulation outcomes define the OAGCNN model performance of 98.5 per cent on the ALDD (apple leaf disease database), with a maximum detection time of 2 seconds. The outcomes validate that the OA-CNN classifier model gives maximum performance for the early-stage detection of ALLD and may execute real-time detection of these diseases with maximum accuracy rate and detection complexity than existing techniques such as Vgg-Icep and multi-label classifier methods.

关键词： Technological innovation image segmentation Databases stochastic processes Production Real-time systems Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Dual-Channel Deep Convolutional Neural Network for Limited-Angle CT image Restoration

Dual-Channel Deep Convolutional Neural Network for Limited-A...

引用

Communication, image and Signal processing (CCISP), International conference on

作者： Yu Chen LingLi Zhang School of Mathematics and Statistics Chongqing Jiaotong University Chongqing China Chongqing Key Laboratory of Statistical Optimization and Complex Data Chongqing University of Arts and Sciences Chongqing China Chongqing Key Laboratory of Group & Graph Theories and Applications Chongqing University of Arts and Sciences Chongqing China

ISBN: (数字)9798350356656

ISBN: (纸本)9798350356663

Based on the excellent performance of computed tomography (CT) in visualizing the inside of objects, it has become one of the indispensable technologies in the fields of medical diagnosis and industrial inspection. However, due to uncontrollable factors such as radiation dose and detection environment, it is necessary to consider how to obtain projection information within a certain scanning rotation angle range in most cases. In these situations, the reconstructed images using traditional analytical algorithms and iterative algorithms will suffer from limited-angle artifacts. Recently, deep learning technology has gained great attention in the field of image restoration due to its powerful learning ability and its rapid development and application. In order to make up for the limitations of traditional methods in suppressing artifacts in limited-angle CT reconstruction, this paper designs a dual-channel deep convolutional network to remove artifacts in reconstructed images with preserving the details and edges. Simulated and real experiments demonstrate that the presented network can effectively restore images, compared with the classical network.

关键词： Visualization Attention mechanisms Convolution Computed tomography Splicing image edge detection Feature extraction image restoration Convolutional neural networks image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Proceedings of the 2022 IEEE International conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2022

Proceedings of the 2022 IEEE International Conference on Ind...

引用

2022 IEEE International conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2022

ISBN: (纸本)9781665451260

The proceedings contain 46 papers. The topics discussed include: fault tolerant scheme in steganographic video streaming using n - repetition code;optimization of approach time waiting at the port of Tanjung Perak Surabaya using genetic algorithm;implementing a low-cost control unit network focusing on data collection and flettner rotor control;a method for designing an embedded human activity recognition system for a kitchen use case based on machine learning;synthetic occluded masked face recognition using convolutional neural networks;energy distribution image processing of stroke EEG signal using gray level co-occurrence matrix;performance comparison of open speech-to-text engines using sentence transformer similarity check with the Korean language by foreigners;a robust offline reinforcement learning algorithm based on behavior regularization methods;a stochastic asynchronous gradient descent algorithm with delay compensation mechanism;human activity recognition from skeletal data using covariance descriptor and temporal subspace clustering;and systematic literature review on process mining in learning management system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

processing and properties of flax fibers reinforced PLA/PBS biocomposites

Processing and properties of flax fibers reinforced PLA/PBS ...

引用

11th conference on Solid State Surfaces and Interfaces

作者： Ketata, Nihel Seantier, Bastien Guermazi, Noamen Grohens, Yves ENIS Lab Genie Materiaux & Environm LGME BP 1173-3038 Sfax 11733038 Tunisia UBS Inst Rech Dupuy Lome IRDL Rue St Maude F-56100 Lorient France

Reinforced plastic materials are widely used in high sophisticated applications. Fiber length has a strong impact on the mechanical properties of composite materials. Being able to reliably and accurately evaluate the residual fiber length distribution of a sample is the key to understand the mechanisms of fiber breakage through experimental investigations, permitting the prediction of properties for flax fiber reinforced components. To optimize the properties of the composites, the greatest average length is sought. The objective is therefore to investigate the effect of the mixing process (single and twin screw extrusion) and the injection shaping process on flax fibers reinforced composites. However, the subsequent image analysis is frequently performed automatically using the fiber length measurement system "FASEP " that can be used to analyze fiber length distributions automatically. After separating the fibers out of the composite material without any damage, and preparing them, statistical analysis can be done. After image processing and analysis with mathematical methods, a complete investigation of the fiber length distribution could be determined. The main results show that a greater concentration of shorter flax fibers was found in the twin screw extruded composite with more than 16% of particles are smaller than 180 mu m when compared against the single screw extruded material. However, volume average results of flax fibers are higher for composite manufactured with single screw extrusion. The control of fiber breakage and final morphology is studied by Scanning Electron Microscopy. As a result of the injection step, the lengths of the fibers have been reduced, compared to the dimensions after extrusion. No reliable correlation between fiber length distribution and mechanical properties, measured with tensile test was found. Also, there is no significant effect of the melt processing types on the tensile properties considering the experimental errors. Co

关键词： Biocomposites PLA/PBS blends Flax fibers Fiber length distribution Mechanical properties

来源：评论

学校读者我要写书评

暂无评论

Signature Forgery Detection and Verification using Deep Learning Techniques

Signature Forgery Detection and Verification using Deep Lear...

引用

International conference on Computing and Networking Technology (ICCNT)

作者： Sri Chakradhar Nossam Rishi Anirudh Katakam Gopa Pulastya Sarada Jayan Department of Computer Science and Engineering Amrita Vishwa Vidyapeetham Bengaluru India Department of Mathematics Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Signature Verification plays a vital role in this current digital era. This work is aimed at studying how signatures can be verified using image processing and Deep Learning methods for forgery detection. We implement Sobel edge detection techniques to allocate the features coming from original signature images and then fake ones, followed by the statistical analysis that inquires the color channels, shape and the average pixel values. Besides that, we do the LSTM and CNN models, which are used to recognize if the given signature is real or forged. LSTM model bases on the finding of flattened images, instead the convolutional layers are being applied for the feature extraction in CNN model. In this process, we plan to create a way to examine images and signatures in order to ensure the safety and higher level of security. We have attained a good accuracy by using the Convolution Neural Network.

关键词： Deep learning statistical analysis Shape image edge detection Feature extraction Forgery Safety Convolutional neural networks Security Long short term memory

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：