检索结果-内蒙古大学图书馆

23rd International Symposium on Quality Electronic Design (ISQED)

作者： Nowshin, Fabiha Yi, Yang Virginia Tech Dept Elect & Comp Engn Blacksburg VA 24061 USA

ISBN: (纸本)9781665494663

Spiking neural networks (SNNs) are artificial neural network models that show significant advantages in terms of power and energy when realizing deep learning applications. However, the data-intensive nature of machine learning applications imposes a challenging problem to neural network implementations in terms of latency, energy efficiency and memory bottleneck. Therefore, we introduce a scalable deep SNN to address the problem of latency and energy efficiency. We integrate a Computing-In-Memory (CIM) architecture built with a fabricated memristor crossbar array to reduce the memory bandwidth in vector-matrix multiplication, a key operation in deep learning. By applying an inter-spike interval (ISI) encoding scheme to the input signals, we demonstrate the spatiotemporal information processing capability of our designed architecture. The memristor crossbar array has an enhanced heat dissipation layer that reduces the resistance variation of the memristors by similar to 30%. We further develop a time-to-first-spike (TTFS) method to classify the outputs. The designed circuits and architecture can achieve very high accuracies with both digit recognition and the MNIST dataset Our architecture can classify handwritten digits while consuming merely 2.9mW of power with an inference speed of 2 mu s/image. Only 2.51pJ of energy per synaptic connection makes it suitable to apply in deep learning accelerators.

关键词： Spiking neural Network Computing-In-Memory LIF Neuron memristor spatiotemporal architecture

来源：评论

学校读者我要写书评

暂无评论

A Mobile App for Detecting Potato Crop Diseases

引用

JOURNAL OF IMAGING 2024年第2期10卷 47-47页

作者： Pineda Medina, Dunia Miranda Cabrera, Ileana de la Cruz, Rolisbel Alfonso Guerra Arzuaga, Lizandra Cuello Portal, Sandra Bianchini, Monica Ctr Nacl Sanidad Agr San Jose De Las Lajas 11300 Cuba Univ Siena Dipartimento Ingn Informaz & Sci Matemat I-53100 Siena Italy

artificial intelligence techniques are now widely used in various agricultural applications, including the detection of devastating diseases such as late blight (Phytophthora infestans) and early blight (Alternaria solani) affecting potato (Solanum tuberorsum L.) crops. In this paper, we present a mobile application for detecting potato crop diseases based on deep neural networks. The images were taken from the PlantVillage dataset with a batch of 1000 images for each of the three identified classes (healthy, early blight-diseased, late blight-diseased). An exploratory analysis of the architectures used for early and late blight diagnosis in potatoes was performed, achieving an accuracy of 98.7%, with MobileNetv2. Based on the results obtained, an offline mobile application was developed, supported on devices with Android 4.1 or later, also featuring an information section on the 27 diseases affecting potato crops and a gallery of symptoms. For future work, segmentation techniques will be used to highlight the damaged region in the potato leaf by evaluating its extent and possibly identifying different types of diseases affecting the same plant.

关键词： deep neural networks image processing classification of potato crop diseases

来源：评论

学校读者我要写书评

暂无评论

Task offloading and resource allocation algorithm based on deep reinforcement learning for distributed AI execution tasks in IoT edge computing environments

引用

COMPUTER networks 2023年第1期223卷 1页

作者： Aghapour, Zahra Sharifian, Saeed Taheri, Hassan Amirkabir Univ Technol Dept Elect Engn Tehran 15914 Iran

Recently, the application of artificial Intelligence (AI) in the Internet of Things (IoT) devices is increasing. As these devices are limited in processing and storing massive computations of AI applications, researchers are searching for methods to overcome these limitations. One of these applications is Convolutional neural Network (CNN) processing, which is common in object detection and image classification. A neural network consists of layers with a large number of neurons and requires high processing power to run. A CNN can be partitioned into segments and offloaded as tasks of IoT devices to cloudlet servers on the edge. By utilizing the edge servers available in the environment, the total latency in the system and energy consumption by IoT devices can be optimized. Making decisions about offloading CNN segmented layers and allocating resources to each of them is the challenge. In this paper, we propose a method based on deep reinforcement learning that divides the offloading and resource allocation problem into two minor problems. This algorithm updates the offloading policy based on information from the environment, and with the help of the Salp Swarm Algorithm (SSA), optimizes resource allocation. The proposed method is tested for different deep-learning tasks of IoT devices under different capacities of cloudlet servers. The simulation results show the proposed algorithm has the least cost in terms of latency and power consumption and on average has improved 92%, 17%, and 12% compared to full local, full offload, and Jointly Resource allocation and computation Offloading PSO (JROPSO) methods respectively.

关键词： Internet of things Edge computing Convolutional neural network Task offloading Resource allocation Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

ViTs are Everywhere: A Comprehensive Study Showcasing Vision Transformers in Different Domain

ViTs are Everywhere: A Comprehensive Study Showcasing Vision...

引用

Cognitive Computing and Complex Data (ICCD), International Conference on the

作者： Md Sohag Mia Abu Bakor Hayat Arnob Abdu Naim Abdullah Al Bary Voban Md Shariful Islam School of Artificial Intelligence Nanjing University of Information Science and Technology Nanjing China Beijing Institute of Technology Beijing China

Transformer design is the de facto standard for natural language processing tasks. The success of the transformer design in natural language processing has lately piqued the interest of researchers in the domain of computer vision. When compared to Convolutional neural networks (CNNs), Vision Transformers (ViTs) are becoming more popular and dominant solutions for many vision problems. Transformer-based models outperform other types of networks, such as convolutional and recurrent neural networks, in a range of visual benchmarks. We evaluate various vision transformer models in this work by dividing them into distinct jobs and examining their benefits and drawbacks. ViTs can overcome several possible difficulties with convolutional neural networks (CNNs). The goal of this survey is to show the first use of ViTs in CV. In the first phase, we categorize various CV applications where ViTs are appropriate. image classification, object identification, image segmentation, video transformer, image denoising, and NAS are all CV applications. Our next step will be to analyze the state-of-the-art in each area and identify the models that are currently available. In addition, we outline numerous open research difficulties as well as prospective research possibilities.

关键词：

来源：评论

学校读者我要写书评

暂无评论

artificial intelligence-powered precision: Unveiling the landscape of liver disease diagnosis-A comprehensive review

引用

ENGINEERING applications OF artificial INTELLIGENCE 2024年 138卷

作者： Vadlamudi, Sireesha Kumar, Vimal Ghosh, Debjani Abraham, Ajith Bennett Univ Sch Comp Sci Engn & Technol TechZone II Greater Noida Uttar Pradesh India Bennett Univ Sch Artificial Intelligence TechZone II Greater Noida Uttar Pradesh India

The global significance of diagnosing liver diseases is heightened, particularly in under-resourced regions with limited healthcare facilities. Traditional diagnostic methods, characterized by time and labor-intensive processes, have led to a growing demand for telemedicine-based solutions. The incorporation of artificial Intelligence is deemed essential to enhance the efficiency and accuracy of diagnostic models. This review explores the seamless integration of diverse data modalities, including clinical records, demographics, laboratory values, biopsy specimens, and imaging data, emphasizing the importance of combining both types for comprehensive liver disease diagnosis. The study rigorously examines various approaches, focusing on pre-processing and feature engineering in non-image data diagnostic model development. Additionally, it analyzes studies employing Convolutional neural networks for cutting-edge solutions in image data interpretation. The paper provides insights into existing liver disease datasets, encompassing both image and non-image data, offering a comprehensive understanding of crucial research data sources. Emphasis is placed on performance evaluation metrics and their correlation in assessing diagnostic model efficiency. The review also explores open-source software tools dedicated to computer-aided liver analysis, enhancing exploration in liver disease diagnostics. Serving as a concise handbook, it caters to novice and experienced researchers alike, offering essential insights, summarizing the latest research, and providing a glimpse into emerging trends, challenges, and future trajectories in liver disease diagnosis.

关键词： Liver diagnosis Telemedicine image & non-image data Pathology Deep learning artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

ViTs are Everywhere: A Comprehensive Study Showcasing Vision Transformers in Different Domain

arXiv

引用

arXiv 2023年

作者： Mia, Md Sohag Arnob, Abu Bakor Hayat Naim, Abdu Al Bary Voban, Abdullah Islam, Md Shariful School of Artificial Intelligence Nanjing University of Information Science and Technology Nanjing China Beijing Institute of Technology Beijing China

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Exploring the Interplay of Interpretability and Robustness in Deep neural networks: A Saliency-Guided Approach

Exploring the Interplay of Interpretability and Robustness i...

引用

image processing Challenges (ICIPC), IEEE International Conference on

作者： Amira Guesmi Nishant Suresh Aswani Muhammad Shafique eBrain Lab Division of Engineering New York University Abu Dhabi Abu Dhabi UAE Department of Computer Science and Engineering New York University Tandon USA

ISBN: (数字)9798331515942

ISBN: (纸本)9798331515959

Adversarial attacks pose a significant challenge to deploying deep learning models in safety-critical applications. Maintaining model robustness while ensuring interpretability is vital for fostering trust and comprehension in these models. This study investigates the impact of Saliency-guided Training (SGT) on model robustness, a technique aimed at improving the clarity of saliency maps to deepen understanding of the model’s decision-making process. Experiments were conducted on standard benchmark datasets using various deep learning architectures trained with and without SGT. Findings demonstrate that SGT enhances both model robustness and interpretability. Additionally, we propose a novel approach combining SGT with standard adversarial training to achieve even greater robustness while preserving saliency map quality. Our strategy is grounded in the assumption that preserving salient features crucial for correctly classifying adversarial examples enhances model robustness, while masking non-relevant features improves interpretability. Our technique yields significant gains, achieving a 35% and 20% improvement in robustness against PGD attack with noise magnitudes of 0.2 and 0.02 for the MNIST and CIFAR-10 datasets, respectively, while producing high-quality saliency maps.

关键词： Training Deep learning image processing Noise Focusing artificial neural networks Robustness Standards Resilience Glass box

来源：评论

学校读者我要写书评

暂无评论

Hybrid Deep Convolutional Generative Adversarial networks (DCGANS) and Style Generative Adversarial Network (STYLEGANS) Algorithms to Improve image Quality 3

Hybrid Deep Convolutional Generative Adversarial Networks (D...

引用

3rd International Conference on Electronics and Sustainable Communication Systems, ICESC 2022

作者： Hariharan, B. Karthic, S. Indra Priyadharshini, S. Nalina, E. Wilfred Blessing, N.R. Senthil Prakash, P.N. Srm Institute of Science and Technology Department of Computational Intelligence Tamil Nadu India Kpr Institute of Engineering and Technology Department of Artificial Intelligence and Data Science Coimbatore India Vellore Institute of Technology School of Computer Science and Engineering Tamil NaduCampus Chennai India R.M.D. Engineering College Tamil Nadu India It University of Technology and Applied Sciences Ibri Oman Rmk College of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu India

ISBN: (数字)9781665479714

ISBN: (纸本)9781665479714

Generative Adversarial networks or GANS, are another way of achieving generative modeling using various deep learning methods like convoluted neural networks. They have a wide range of applications like :- image to image translation, improving the resolution of the images, creating multiple images using a single image, checking if the image is real or fake and the list goes on. One of the implementations of GANS, i.e, DCGANS would be discussed in this paper. DCGANS stands for Deep Convolutional Generative Adversarial networks, it uses convolutional and convolutional-transpose layers in the generator and discriminator, respectively. This DCGANS give prompt result while processing many real time images. In their study Unsupervised representation Learning with Deep Convolutional Generative Adversarial networks, Radford et al. proposed it. This algorithm generates fake images after being trained on a dataset of human faces. However to create these fake images, the quality of these images should be compromised and this can also make them easily identifiable as fake image. So, here StyleGANS will be used as another implementation of GANS. StyleGANS is not only used for improving the quality of these images but also it helps us to increase fakeness of the images. This project uses the CelebA dataset from Kaggle to train the models. © 2022 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Spiking neural networks for Classification of Brain-Computer Interface and image Data

Spiking Neural Networks for Classification of Brain-Computer...

引用

2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

作者： Honzik, Vaclav Moucek, Roman University of West Bohemia Department of Computer Science and Engineering Plzeň306 14 Czech Republic

ISBN: (纸本)9781665401265

Spiking neural networks are a promising concept not only in terms of better simulation of biological neural networks but also in terms of overcoming the current disadvantages of artificial neural networks, such as high energy consumption or slow response time. The paper focuses on the potential benefits of spiking neural networks in the classification of event-related components processed in many traditional brain-computer interface experiments. Experiments with various spiking network architectures and optimization approaches over specific brain computer interface and image datasets are presented, and their results are provided and discussed. The best accuracy achieved was 64.86% for the event-related component dataset and 97.09% for the image dataset. © 2021 IEEE.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

Classification and tracking of items on a moving conveyor belt using convolutional networks and image processing

Classification and tracking of items on a moving conveyor be...

引用

artificial Intelligence, Computer, Data Sciences and applications (ACDSA), International Conference on

作者： Angelo Leite Medeiros De Góes Adrião Duarte Dória Neto Matheus Gomes Diniz Andrade Thiago Theiry De Oliveira Electrical and Computer Engineering Postgraduate Program Federal University of Rio Grande do Norte Natal Brazil Computer Engineering Federal University of Rio Grande do Norte Natal Brazil

The prolonged waiting time at supermarket checkout lines poses a significant challenge to the shopping experience, impacting customer satisfaction and operational efficiency. This paper presents a prototype that addresses this issue by utilizing computer vision and deep learning. The model, incorporating convolutional neural networks such as YOLO v4 tiny and YOLO v5 small, along with tools like OpenCV and Roboflow for dataset management, achieves a remarkable 98% mean average precision for two-class detection. It efficiently detects, classifies, tracks, and counts items on a mobile supermarket conveyor belt. Additionally, we introduce a versatile framework designed for seamless integration into real-world applications. It comprises a customizable monitoring application and simulator that facilitates synthetic image data generation. Managing diverse items in a supermarket presents a major challenge for data gathering, labeling, and training. In that sense, the importance of customizable monitoring and simulation tools is highlighted, emphasizing their practical role. Our findings demonstrate the feasibility of maintaining a minimal 0% to 2.85% precision tradeoff while using half of the data as synthetic for two-class detection, indicating potential practicality in supermarkets with proper scaling. In summary, this study brings tangible benefits to both customers and retailers, offering a potential to streamline, speed up, and cut costs in the supermarket checkout process.

关键词：

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：