检索结果-内蒙古大学图书馆

applications of deep learning in fish habitat monitoring: A tutorial and survey

EXPERT SYSTEMS WITH applications 2024年第PartB期238卷

作者： Saleh, Alzayat Sheaves, Marcus Jerry, Dean Azghadi, Mostafa Rahimi James Cook Univ Coll Sci & Engn 1 James Cook Dr Townsville Qld 4811 Australia James Cook Univ ARC Res Hub Supercharging Trop Aquaculture Genet S 1 James Cook Dr Townsville Qld 4811 Australia

Marine ecosystems and their fish habitats are becoming increasingly important due to their integral role in providing a valuable food source and conservation outcomes. Due to their remote and difficult to access nature, marine environments and fish habitats are often monitored using underwater cameras to record videos and images for understanding fish life and ecology, as well as for preserve the environment. There are currently many permanent underwater camera systems deployed at different places around the globe. In addition, there exists numerous studies that use temporary cameras to survey fish habitats. These cameras generate a massive volume of digital data, which cannot be efficiently analysed by current manual processing methods, which involve a human observer. Deep Learning (DL) is a cutting-edge artificial Intelligence (AI) technology that has demonstrated unprecedented performance in analysing visual data. Despite its application to a myriad of domains, its use in underwater fish habitat monitoring remains under explored. In this paper, we provide a tutorial that covers the key concepts of DL, which help the reader grasp a high-level understanding of how DL works. The tutorial also explains a step-by-step procedure on how DL algorithms should be developed for challenging applications such as underwater fish monitoring. In addition, we provide a comprehensive survey of key deep learning techniques for fish habitat monitoring including classification, counting, localisation, and segmentation. Furthermore, we survey publicly available underwater fish datasets, and compare various DL techniques in the underwater fish monitoring domains. We also discuss some challenges and opportunities in the emerging field of deep learning for fish habitat processing. This paper is written to serve as a tutorial for marine scientists who would like to grasp a high-level understanding of DL, develop it for their applications by following our step-by-step tutorial, and see how

关键词： Marine Science Computer Vision Convolutional neural networks image and video processing Machine Learning Deep Learning Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

image Segmentation Losses with Modules Expressing a Relationship Between Predictions 19th

Image Segmentation Losses with Modules Expressing a Relati...

引用

19th International Conference on Information processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2022

作者： Hurtik, Petr Molek, Vojtech Zámečníková, Hana University of Ostrava Centre of Excellence IT4Innovations Institute for Research and Applications of Fuzzy Modeling 30. dubna 22 Ostrava Czech Republic

ISBN: (纸本)9783031089732

We focus on semantic image segmentation with the usage of deep neural networks and give emphasis on the loss functions used for training the networks. Considering region-based losses, Dice loss, and Tversky loss, we propose two independent modules that easily modify the loss functions to take into account the relationship between the class predictions and increase the slope of the gradient. The first module expresses the ambiguity between classes and the second module utilizes a differentiable soft argmax function. Each of the modules is used before the standard loss is computed and remains untouched. In the benchmark, we involved two neural network architectures with two different backbones, selected two loss functions, and examined separately two scenarios for softmax and sigmoid top activation functions. In the experiment, we demonstrate the usefulness of our modules by improving the IOU and F1 coefficients on the test dataset for all scenarios tested. Moreover, the usage of the modules decreases overfitting. The proposed modules are easy to integrate into existing solutions and add near-zero computational overhead. © 2022, Springer Nature Switzerland AG.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on Computer Deep Learning Algorithm to Optimize the Fault Prediction Model of Nuclear Power Plant Emergency Feed Pump

Research on Computer Deep Learning Algorithm to Optimize the...

引用

image processing and Computer applications (ICIPCA), IEEE International Conference on

作者： Chao Gao Xianjun Song Zhiqiang Zhang Chunguang Li Xincai Chang CGNPC CGN Digital Technology Co. LTD Beijing China

ISBN: (数字)9798350360240

ISBN: (纸本)9798350384161

This project takes the typical failure forms of pumps in nuclear power plants such as abnormal vibration, friction and wear as the research object. The most readily available pump housing acceleration signal frequency domain information is used as input. The system is mainly composed of four modules: input module, convolution module, multi-head self-attention module and output module. The attention mechanism of characteristic frequency domain data based on the fusion of deep neural network and attention network is studied. At the same time, the pattern recognition model of pump faults in nuclear power plant is constructed. Compared with the existing research results, using frequency-domain data as input, using frequency-domain data attention network and other technologies can shorten the amount of input data, and make the recognition accuracy of the model up to 100% in the test set, which has obvious advantages over other fault recognition models based on deep neural networks.

关键词： Vibrations Accuracy Frequency-domain analysis Computational modeling artificial neural networks Mathematical models Data models Water pumps Feeds Power generation

来源：评论

学校读者我要写书评

暂无评论

Wheat Diseases Detection and Classification using Convolutional neural Network (CNN)

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND applications 2022年第11期13卷 719-726页

作者： Hossen, Md Helal Mohibullah, Md Muzammel, Chowdhury Shahriar Ahmed, Tasniya Acharjee, Shuvra Panna, Momotaz Begum Comilla Univ Dept CSE Cumilla Bangladesh Noakhali Sci & Technol Univ Inst Informat Technol Noakhali Bangladesh Jahangirnagar Univ Dept CSE Dhaka Bangladesh

Ever since the medieval era, the preponderance of our concentration has been concentrated upon agriculture, which is typically recognized to be one of the vital aspects of the economy in contemporary society. This focus on agriculture can be traced back to the advent of the industrial revolution. Wheat is still another type of grain that, in the same way as other types of harvests, satisfies the necessity for the essential nutrients that are required for our bodies to perform their functions correctly. On the other hand, the supply of this harvest is being limited by a variety of rather frequent ailments. This is making it difficult to meet demand. The vast majority of people who work in agriculture are illiterate, which hinders them from being able to take appropriate preventative measures whenever they are necessary to do so. As a direct consequence of this factor, there has been a reduction in the total amount of wheat that has been produced. It can be quite difficult to diagnose wheat illnesses in their early stages because there are so many various forms of environmental variables and other factors. This is because there are numerous distinct sorts of agricultural products, illiteracy of agricultural workers, and other factors. In the past, a variety of distinct models have been proposed as potential solutions for identifying illnesses in wheat harvests. This study demonstrates a two-dimensional CNN model that can identify and categorize diseases that affect wheat harvests. To identify significant aspects of the photos, the software employs models that have previously undergone training. The suggested method can then identify and categorize disease-affected wheat crops as distinct from healthy wheat crops by employing the major criteria described above. The reliability of the findings was assessed to be 98.84 percent after the collection of a total of 4800 images for this study. These images included eleven image classes of images depicting diseased crops and o

关键词： Wheat crop diseases artificial intelligence convolution neural networks image processing feature extraction

来源：评论

学校读者我要写书评

暂无评论

A hybrid convolutional architecture for accurate image manipulation localization at the pixel-level

引用

MULTIMEDIA TOOLS AND applications 2021年第15期80卷 23377-23392页

作者： Zhang, Yixuan Zhang, Jiguang Xu, Shibiao Univ Chinese Acad Sci Sch Artificial Intelligence Beijing Peoples R China Chinese Acad Sci Inst Informat Engn State Key Lab Informat Secur Beijing Peoples R China Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing Peoples R China

Advanced image processing techniques can easily edit images without leaving any visible traces, making manipulation detection and localization for forensics analysis a challenging task. Few studies can simultaneously locate tampered objects accurately and refine contours of tampered regions effectively. In this study, we propose an effective and novel hybrid architecture, named Pixel-level image Tampering Localization Architecture (PITLArc), which integrates the advantages of top-down detection-based methods and bottom-up segmentation-based methods. Moreover, we provide a typical fusion implementation of our proposed hybrid architecture on one outstanding detection-based method (two-stream faster region-based convolutional neural network (RGB-N)) and two segmentation-based methods (Multi-Scale Convolution neural networks (MSCNNs) and Dual-domain Convolutional neural networks (DCNNs)) to evaluate the effectiveness of the proposed architecture. The three methods can be integrated into our proposed PITLArc to significantly improve their performance. Other detection and segmentation algorithms (not limited to the three aforementioned methods) can also be integrated into our architecture to improve their performance. Moreover, a Dense Conditional Random Fields (DenseCRFs)-based post-processing method is introduced to further optimize the details of tampered regions. Experiments validate the effectiveness of the proposed architecture.

关键词： Manipulation localization Top-down detection Bottom-up segmentation DenseCRFs

来源：评论

学校读者我要写书评

暂无评论

image processing Based Lie Detector Model

Image Processing Based Lie Detector Model

引用

Innovations in Intelligent Systems and applications Conference

作者： Aybüke Akyakı Büşra Karaca Doğay Altınel Filiz Gürkan Elektrik-Elektronik Mühendisliği Bölümü İstanbul Medeniyet Üniversitesi İstanbul Türkiye

ISBN: (数字)9798350379433

ISBN: (纸本)9798350379440

image processing refers to information processing techniques used with computers integrated into images in various industrial and technological fields. These techniques can be used in various applications involving the process of object recognition and are developed within the scope of artificial intelligence. In recent years, image processing techniques have been utilized in facial recognition systems to detect crime and criminals. This paper presents a study on the creation of a lie detector that provides information on whether individuals are lying by using image processing techniques. For this purpose, algorithms that analyze eye blinking and emotional state behaviors from the facial region detected in the video image for lie detection were developed. These algorithms were coded in the Python programming language and combined to obtain a lie detector model. Additionally, a balanced test set consisting of video images of individuals telling the truth and lying were prepared to validate the model. In tests conducted on this set, the success rate of the obtained lie detector model was found to be 83.3%.

关键词： Ear image processing Detectors Brain modeling Python Ubiquitous computing Technological innovation Skin Object recognition neural networks

来源：评论

学校读者我要写书评

暂无评论

A design and development of support system for prediction of various renal syndromes using artificial neural networks

引用

INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT 2021年 1-16页

作者： Sumana, Gollapalli Kalaiselvi, K. Vijayalakshmi, J. Shanthi, S. Aparna, G. Joseph, M. Kezia Sri Padmavati Mahila Visvavidyalayam Command Control Ctr NTR Marg Tirupati Andhra Pradesh India SRM Inst Sci & Technol Dept Networking & Commun Chennai Tamil Nadu India Sri Sai Ram Engn Coll IT Dept Chennai Tamil Nadu India Malla Reddy Coll Engn & Technol Dept Comp Sci & Engn Hyderabad India Osmania Univ Univ Coll Engn Dept Elect & Commun Engn Hyderabad 500040 Telangana India Stanley Coll Engn & Technol Women Dept ECE Hyderabad Telangana India

The core intention of this theory is prognosis of differential renal syndromes and prediction of nephrolithiasis utilizing artificial neural networks (ANN). An artificial neural Network (ANN) is an information processing paradigm that is propelled similar to the natural sensory organs, such as the brain, nerves that process, exchange and stores the data. This system is produced to support the medical specialist in their determination by enabling them to less utilization of time in recognizing small imperfections in the Renal system. It is fundamental to remind that this product does not affirm to supplant the physician, as expert evaluation assessment is regularly essential, particularly in new cases. The key component of this proposed pattern is the new structure of the information processing system. neural networks, because of its striking capacity to obtain unpredictable or uncertain information, is applied to extract patterns and perceive trends that, are too hard to handle to the human beings or other computer techniques [Kevin J Am Soc Nephrol, 2019]. The developed and trained neural network is considered as an "expert" in the class of information that is provided to analyze. This expert system then can be used to afford projections given new patterns. neural networks are widely used in the areas like character recognition, image compression, stock market prediction, travelling sale's man problem;security based miscellaneous and medical applications. In other words the conventional technique for the human subjective assessment strategy is being substituted by computer based systems utilizing soft computing approaches [Am Soc Nephrol, 2019]. Presently neural networks are widely utilized as a part of medical fields so that the proposed system is designed, constructed and developed for diagnosis of differential renal syndromes that acts as most excellent support for the therapeutic professionals [Xie Kidney Dis, 2020][Dayyal Bioscience, 2018]. As there is brisk i

关键词： Renal syndromes are fundamentally analyzed in five stages These stages depend on physical symptoms Urinary complications Dipstick test evaluation of the urine sample Clinical evaluation of urine sample and blood sample of a patient

来源：评论

学校读者我要写书评

暂无评论

Power Efficient Real-Time Traffic Signal Classification for Autonomous Driving Using FPGAs

Power Efficient Real-Time Traffic Signal Classification for ...

引用

International Conference on Communications, Signal processing, and their applications (ICCSPA)

作者： Rashed Al Amin Md Shahi Amran Hossain Leon Tjard Schmid Veit Wiese Roman Obermaisser Institute for Embedded Systems University of Siegen Siegen Germany

ISBN: (数字)9798350384819

ISBN: (纸本)9798350384826

The significant advancements in research within the domain of Advanced Driver-Assistance Systems (ADAS) have been significantly propelled by the integration of artificial Intelligence (AI) and Deep Learning (DL). The detection and classification of traffic signals are pivotal for ensuring driver safety and comfort, prompting the development of numerous detection and classification methodologies in recent times. Convolutional neural networks (CNNs) and Field Programmable Gate Arrays (FPGAs) have individually demonstrated notable efficacy in signal processing, image detection, and image classification tasks. However, the amalgamation of CNNs and FPGAs for traffic signal classification offers enhanced flexibility, accuracy, and resource efficiency. This study proposes a software-hardware co-design approach for the development of a resource-efficient real-time traffic signal classification system tailored for FPGAs utilizing the CNN algorithm. The software implementation has been carried out within the TensorFlow framework, while the hardware implementation is executed via the Xilinx Vivado 2023.1 platform and rigorously evaluated using the Xilinx ZCU102 FPGA board. Experimental findings demonstrate that the traffic signal classification accuracy can attain approximately 99.98% at a frame rate of 84139 frames per second (FPS) on the FPGA platform, with a power consumption of 4.4W. The exceptionally low latency and heightened accuracy exhibited by the proposed system render it highly advantageous for ADAS and other safety-critical applications.

关键词： Accuracy Software algorithms Pattern classification Signal processing algorithms Real-time systems Software Safety Convolutional neural networks artificial intelligence Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

Deep Spiking neural networks Based on STDP Rules for image Recognition

Deep Spiking Neural Networks Based on STDP Rules for Image R...

引用

International Conference on Wavelet Active Media Technology and Information processing (ICWAMTIP)

作者： Can Yan Guang-Shuo Wang Jian-Ping Li School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

Spiking neural networks (SNN) can transmit and process information through a sequence of spikes. Compared with the first two generations of artificial neural networks, spiking neural networks are closer to biological neurons and have high biological interpretability. In addition, SNN have low power consumption and good applications in neurochips. SNN learning algorithms include unsupervised algorithms and supervised algorithms. The STDP rule, as an archetype of unsupervised algorithms, is widely used and has been derived from several versions. This review concentrates on the application of STDP algorithm in the areas of image recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Spike-Event-Driven Deep Spiking neural Network With Temporal Encoding

引用

IEEE SIGNAL processing LETTERS 2021年 28卷 484-488页

作者： Zhang, Zhixuan Liu, Qi Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu 610054 Peoples R China Natl Univ Singapore Dept Elect & Comp Engn Singapore 117583 Singapore

Feature extractionplays an important role before pattern recognition takes place. The existing artificial neural networks (ANNs), however, ignoreto learn and represent temporal information, instead of only utilizing spatial information for recognition. Moreover, the substantial computational and energy costs resulted from the conventional ANN-based classifiers, limit their uses in mobile and embedded applications. In this work, we develop a sparse temporal encoding method which exploits both spatial and temporal information. On the basis of spike-timing-dependent plasticity and multi-scale structure, the resulting temporal feature representation integrates with a temporal spiking neural network (SNN) classifier to achieve high efficiency of parallel computing for feature extraction. Experimental evaluation on four benchmark datasets from image classification and speech recognition tasks show the proposed SNN model yielding state-of-the-art accuracy.

关键词： Neurons Encoding Feature extraction Computational modeling Task analysis image coding Biological neural networks image classification multi-scale speech recognition spike-event-driven spiking neural network

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：