Muutke küpsiste eelistusi

E-raamat: Computer Vision and Recognition Systems Using Machine and Deep Learning Approaches: Fundamentals, technologies and applications

Edited by (University of Missouri at Saint Louis, Department of Computer Science, USA-), Edited by (Charles Darwin University, College of Engineering, IT, and Environment, Australia), Edited by , Edited by (VIT University, School of Information Technology and Engineering, India), Edited by
  • Formaat: EPUB+DRM
  • Sari: Computing and Networks
  • Ilmumisaeg: 28-Oct-2021
  • Kirjastus: Institution of Engineering and Technology
  • Keel: eng
  • ISBN-13: 9781839533242
Teised raamatud teemal:
  • Formaat - EPUB+DRM
  • Hind: 227,50 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
Computer Vision and Recognition Systems Using Machine and Deep Learning Approaches: Fundamentals, technologies and applicationsSuurem pilt
  • Formaat: EPUB+DRM
  • Sari: Computing and Networks
  • Ilmumisaeg: 28-Oct-2021
  • Kirjastus: Institution of Engineering and Technology
  • Keel: eng
  • ISBN-13: 9781839533242
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

Computer vision is an interdisciplinary scientific field that deals with how computers obtain, store, interpret and understand digital images or videos using artificial intelligence based on neural networks, machine learning and deep learning methodologies. They are used in countless applications such as image retrieval and classification, driving and transport monitoring, medical diagnostics and aerial monitoring.

Written by a team of international experts, this edited book covers the state-of-the-art of advanced research in the fields of computer vision and recognition systems from fundamental concepts to methodologies and technologies and real world applications including object detection, biometrics, Deepfake detection, sentiment and emotion analysis, traffic enforcement camera monitoring, vehicle control and aerial remote sensing imagery.

The book will be useful for industry and academic researchers, scientists and engineers in the fields of computer vision, machine vision, image processing and recognition, multimedia, AI, machine and deep learning, data science, biometrics, security, and signal processing. It will also make a great course reference for advanced students and lecturers in these fields of research.



Written by a team of International experts, this edited book covers state-of-the-art research in the fields of computer vision and recognition systems from fundamental concepts to methodologies and technologies and real-world applications. The book will be useful for industry and academic researchers, scientists and engineers.

About the editors xv
Preface xvii
1 Computer vision and recognition-based safe automated systems
1(16)
Chiranji Lai Chowdhary
Harpreet Kaur
Dharm Singh Jat
Abhishek Ranjan
1.1 Introduction
2(2)
1.1.1 Role of computer vision in automation
3(1)
1.1.2 Organization of the chapter
3(1)
1.2 Literature survey of safe automation systems
4(1)
1.3 Application of computer vision technology in automation
5(6)
1.3.1 Using face ID in mobile devices
6(1)
1.3.2 Automated automobiles
6(1)
1.3.3 Computer vision in agriculture
7(1)
1.3.4 Computer vision in the health sector
8(1)
1.3.5 Computer vision in the e-commerce industry
8(2)
1.3.6 Generating 3D maps
10(1)
1.3.7 Classifying and detecting objects
10(1)
1.3.8 Congregation data for training algorithms
10(1)
1.3.9 Low-light mode with computer vision
10(1)
1.4 Ensuring safety during COVID-19 using computer vision
11(1)
1.4.1 AI started from bringing humans closer to forcing them in keeping apart
11(1)
1.4.2 Access control through computer vision
11(1)
1.4.3 Thermal fever detection cameras
11(1)
1.4.4 Social distancing detection
11(1)
1.4.5 Sanitization prioritization
12(1)
1.4.6 Face mask compliance
12(1)
1.5 Discussion and conclusion
12(5)
References
13(4)
2 DLA: deep learning accelerator
17(34)
Seyedeh Yasaman Hosseini Mirmahaleh
Midia Reshadi
2.1 Introduction
18(1)
2.2 ASIC-based design accelerator
19(4)
2.3 FPGA-based design accelerator
23(2)
2.4 NoC-based design accelerator
25(7)
2.5 Flow mapping and its impact on DLAs' performance
32(8)
2.6 A heuristic or dynamic algorithm's role on a DLA's efficiency
40(4)
2.7 Brief state-of-the-art survey
44(7)
References
46(5)
3 Intelligent image retrieval system using deep neural networks
51(32)
Shubham Gujar
Rutuparn Pawar
Yogesh Dandawate
3.1 Introduction
52(2)
3.2 Conventional content-based image retrieval (CBIR) system
54(2)
3.2.1 Semantic-based image retrieval (SBIR) system
56(1)
3.3 Deep learning
56(2)
3.4 Image retrieval using convolutional neural networks (CNN)
58(9)
3.5 Image retrieval using autoencoders
67(7)
3.6 Image retrieval using generative adversarial networks (GAN)
74(9)
References
79(4)
4 Handwritten digits recognition using dictionary learning
83(34)
Vahid Abolghasemi
Rasoul Ameri
Kianoush Nazarpour
4.1 Introduction
84(3)
4.1.1 Optical character recognition
84(1)
4.1.2 Handwritten recognition
85(2)
4.2 Related works
87(2)
4.3 Dictionary learning
89(3)
4.4 DPL variants for HNR
92(7)
4.4.1 Dictionary pair learning model
93(1)
4.4.2 Incoherent dictionary pair learning (InDPL)
94(2)
4.4.3 Labeled projective dictionary pair learning
96(3)
4.5 Input data preparation
99(2)
4.5.1 Image preprocessing
99(1)
4.5.2 Histogram of oriented gradient
99(2)
4.5.3 Classification stage
101(1)
4.6 HNR datasets
101(1)
4.7 Experimental results
102(8)
4.7.1 Cross-validation
102(4)
4.7.2 Benchmarking results
106(4)
4.8 Conclusions
110(7)
References
111(6)
5 Handwriting recognition using CNN and its optimization approach
117(28)
Phattharaphon Romphet
Supasit Kajkamhaeng
Chantana Chantrapornchai
5.1 Introduction
118(2)
5.2 Related works
120(2)
5.3 Background
122(4)
5.3.1 Convolutional neural network
122(1)
5.3.2 Gated convolutional neural network
122(1)
5.3.3 Gated recurrent unit (GRU)
123(1)
5.3.4 Connectionist temporal classification (CTC)
123(1)
5.3.5 Residual operation
124(1)
5.3.6 Bi-directional gated recurrent unit (BiGRU)
124(1)
5.3.7 Squeeze and excited network (SENet)
124(1)
5.3.8 Linear bottleneck network
125(1)
5.3.9 Encoder and decoder model
125(1)
5.4 Methodology
126(7)
5.4.1 Data gathering
126(1)
5.4.2 Preprocessing
126(1)
5.4.3 Model overview
127(5)
5.4.4 Metrics
132(1)
5.4.5 Training configurations
132(1)
5.4.6 Unseen testing
132(1)
5.4.7 Inference time testing
132(1)
5.4.8 Visualize inside the model
133(1)
5.5 Experiments
133(8)
5.5.1 Experiment 1: Bluche versus Puigcerver versus Flor model
133(1)
5.5.2 Experiment 2: performance comparison of the encoder
134(1)
5.5.3 Experiment 3: performance comparison of the decoder
135(1)
5.5.4 Experiment 4: performance comparison of the skipped connection
136(2)
5.5.5 Experiment 5: performance comparison of other ResFlor model
138(1)
5.5.6 Experiment 6: ResFlor residual with SE network
139(1)
5.5.7 Experiment 7: ResFlor with residual and bottleneck network
140(1)
5.6 Summary
141(1)
5.7 Conclusion and future work
142(3)
Acknowledgments
142(1)
References
142(3)
6 Real-time face mask detection on edge IoT devices
145(20)
Aditya Dinesh Oke
Saumya Verma
Ayush Sinha
K. Deepa
6.1 IoT devices and object detection
145(2)
6.1.1 IoT devices and object detection
145(1)
6.1.2 Real-time object detection on edge IoT devices
146(1)
6.1.3 A generic detection algorithm
146(1)
6.2 Literature survey
147(1)
6.3 Traditional feature extraction techniques
147(2)
6.3.1 Histogram of oriented gradients (HOG)
148(1)
6.3.2 Scale invariant feature transform (SIFT)
148(1)
6.3.3 Speeded up robust features (SURF)
149(1)
6.4 Traditional detection methods
149(1)
6.4.1 Histogram of oriented gradients with support vector machines (HOG + SVM)
149(1)
6.5 Traditional face detection techniques
150(2)
6.5.1 Viola-Jones Haar cascade method
150(2)
6.6 Face mask detection
152(1)
6.7 Deep learning for object detection
152(4)
6.7.1 Convolutional neural networks (CNNs)
152(1)
6.7.2 Object detection using deep learning
153(1)
6.7.3 Faster RCNN for object detection
154(1)
6.7.4 Enhancing faster RCNN with MobileNet
155(1)
6.8 Internet and deep learning
156(2)
6.8.1 Client-server architecture
156(2)
6.9 Edge IoT architecture
158(1)
6.10 Implementing an edge IoT architecture
158(3)
6.10.1 Dynamic web pages
158(1)
6.10.2 Backend using Node.js
159(2)
6.10.3 MongoDB as database
161(1)
6.11 Discussion
161(1)
6.12 Conclusion
161(4)
References
162(3)
7 Current challenges and applications of DeepFake systems
165(18)
Sandeep Bhat
Zeel Naman Shah
G.M. Siddesh
7.1 Introduction to DeepFake
165(1)
7.1.1 Scenario
166(1)
7.2 Various DeepFake detection methods available and their limitations
166(8)
7.2.1 Traditional detection methods
167(4)
7.2.2 Methods based on deep learning
171(3)
7.3 Applications used to forge the multimedia
174(1)
7.4 Current challenges and future of the technology
175(2)
7.4.1 Quality of DeepFake dataset
175(1)
7.4.2 Performance evaluation
176(1)
7.4.3 Explainability of detection results
176(1)
7.4.4 Temporal aggregation
176(1)
7.4.5 Social media laundering
176(1)
7.5 Conclusion
177(6)
References
178(5)
8 Vehicle control system based on eye, iris, and gesture recognition with eye tracking
183(20)
Harpreet Kaur
Chiranji Lai Chowdhary
8.1 Introduction
183(1)
8.2 Eye tracking
184(4)
8.2.1 How eye tracker works
185(3)
8.3 Human gesture
188(3)
8.3.1 Head movement
190(1)
8.4 Applications of eye tracking
191(4)
8.5 Top eye tracking hardware companies
195(1)
8.6 Case studies
196(2)
8.7 Conclusion
198(5)
References
199(4)
9 Sentiment analysis using deep learning
203(20)
Parul Gandhi
Surbhi Bhatia
Norah Alkhaldi
9.1 Sentiment analysis: an interesting problem
204(1)
9.2 Sentiment and opinions
205(1)
9.3 Components of opinion
205(3)
9.3.1 Levels in sentiment analysis
206(1)
9.3.2 Classification techniques
207(1)
9.3.3 Classification types
207(1)
9.4 Deep learning
208(1)
9.5 Machine learning
209(1)
9.6 Traditional learning
210(1)
9.7 Hybrid learning approaches
210(1)
9.8 Deep neural networks
211(1)
9.8.1 Deep belief network
211(1)
9.8.2 Convolutional neural networks
211(1)
9.8.3 Stacked autoencoders
211(1)
9.9 Convolutional neural networks
212(2)
9.9.1 Word embeddings
212(1)
9.9.2 Bag of words (BOW)
213(1)
9.9.3 ConvNet structure
213(1)
9.10 Proposed model
214(5)
9.10.1 Datasets and experimental setup
215(1)
9.10.2 Results
216(1)
9.10.3 Effect of filter region size
217(1)
9.10.4 Effect of number of filters
217(1)
9.10.5 Effect of different classifiers
218(1)
9.11 Conclusions and future scope
219(4)
References
220(3)
10 Classification of prefeature extracted images with deep convolutional neural network in facial emotion recognition of vehicle driver
223(30)
Ganesan Kaliyaperumal
Manikandan N.S.
10.1 Introduction and related work
223(3)
10.2 Proposed models
226(9)
10.2.1 Datasets
226(1)
10.2.2 Preprocessing
227(1)
10.2.3 Prefeature extraction
227(3)
10.2.4 Convolutional neural networks
230(2)
10.2.5 Model design
232(2)
10.2.6 Metrics
234(1)
10.2.7 System configuration
235(1)
10.3 Experiments and results
235(12)
10.4 Vehicle driver emotion recognition experimental setup, results, and discussion
247(2)
10.5 Conclusion
249(4)
Acknowledgments
249(1)
References
249(4)
11 MobileNet architecture and its application to computer vision
253(24)
Rupa Patel
Anita Chaware
11.1 Introduction
254(1)
11.2 Preliminaries
255(7)
11.2.1 Artificial neural network
255(3)
11.2.2 Convolution neural network
258(4)
11.2.3 Deep convolution neural network
262(1)
11.3 Benchmarked convolutional neural network
262(1)
11.3.1 VGG16
262(1)
11.3.2 Inception v3
262(1)
11.4 MobileNet architecture
262(5)
11.4.1 MobileNetvl
262(2)
11.4.2 MobileNetv2
264(2)
11.4.3 MobileNetv3
266(1)
11.4.4 NASNet mobile
266(1)
11.5 Model optimization techniques
267(1)
11.5.1 Quantization technique
267(1)
11.6 Quantized deep convolutional neural network
267(2)
11.6.1 Methodology
268(1)
11.7 Case study: healthcare domain
269(2)
11.7.1 Diabetic retinopathy
269(1)
11.7.2 Kaggle diabetic retinopathy image datasets
269(1)
11.7.3 Approach
270(1)
11.7.4 Experiment results and discussion
270(1)
11.7.5 Conclusion
270(1)
11.8 Selected MobileNet application
271(2)
11.8.1 Image classification
271(1)
11.8.2 Object detection
272(1)
11.8.3 Segmentation
273(1)
11.9 Future direction
273(4)
References
273(4)
12 Study on traffic enforcement cameras monitoring to detect the wrong-way movement of vehicles using deep convolutional neural network
277(24)
S.R. Mani Sekhar
Sainya Goyal
Lakshya Aditi Sinha
G.M. Siddesh
12.1 Introduction
277(1)
12.2 Background
278(1)
12.3 Techniques for data collection
279(1)
12.3.1 Closed-circuit television
279(1)
12.3.2 Manual videos
279(1)
12.4 Purpose and benefit of the cameras monitoring system
280(1)
12.5 Techniques used in the monitoring of vehicles
281(6)
12.5.1 Convolution neural network
281(3)
12.5.2 R-CNN
284(1)
12.5.3 Fast region-based convolution neural network
285(1)
12.5.4 Faster R-CNN
285(1)
12.5.5 Single-shot MultiBoxDetector
286(1)
12.5.6 You Only Look Once
286(1)
12.6 Case study
287(8)
12.6.1 The detection of wrong-way drive of automobiles based on appearance using deep convolutional neural network
287(2)
12.6.2 Real-time wrong-direction detection based on deep learning
289(2)
12.6.3 A vehicle finding and counting system based on vision using deep learning
291(2)
12.6.4 A highway automobile discovery algorithm based on CNN
293(2)
12.6.5 Comparison of case studies
295(1)
12.7 Conclusion
295(6)
References
297(4)
13 Glasses for smart tourism applications
301(36)
Soorya Ram Shimgekar
Pathi Preetham Reddy
Praveen Kumar Reddy Maddikunta
Giridhar Reddy Bojja
13.1 Introduction
302(3)
13.1.1 Motivation
304(1)
13.1.2 Contribution of our work
305(1)
13.2 Article structure
305(2)
13.3 Existing technologies related to smart glasses
307(6)
13.3.1 Applications of smart glasses
307(1)
13.3.2 Smart glasses technology in the market
307(1)
13.3.3 Smart glasses solutions papers
307(6)
13.4 System assumptions
313(1)
13.5 Functional architecture and technologies relevant
314(6)
13.5.1 Voice to text conversion-KALDI
314(2)
13.5.2 Name remember
316(1)
13.5.3 Facial features extraction
317(1)
13.5.4 Object (plant and animal) identification
317(1)
13.5.5 Text detection
318(1)
13.5.6 Text-translation
319(1)
13.5.7 Navigation
319(1)
13.5.8 Text to speech conversion
320(1)
13.6 Proposed style of interaction (KBSIS)
320(5)
13.6.1 KALDI
320(1)
13.6.2 Dataset and model used
321(1)
13.6.3 Inputs
322(1)
13.6.4 Inference from input and processing
322(1)
13.6.5 Outputs
323(2)
13.7 Results and discussion
325(5)
13.7.1 Navigation
325(1)
13.7.2 Music
325(1)
13.7.3 Text from image and Translate
325(1)
13.7.4 Remembering face and naming
325(2)
13.7.5 Face characteristics
327(1)
13.7.6 Weather
328(1)
13.7.7 Plant identification and search
328(1)
13.7.8 Animal identification and search
328(2)
13.8 Conclusion
330(7)
References
330(7)
14 Renal calculi detection using modified grey wolf optimization
337(14)
Isha Sharma
Vijay Kumar
14.1 Introduction
337(1)
14.2 Background
338(5)
14.2.1 Image segmentation
338(1)
14.2.2 Grey wolf optimization
339(2)
14.2.3 Previous work
341(2)
14.3 Proposed approach for renal calculi detection
343(3)
14.3.1 Challenges in renal calculi detection
343(1)
14.3.2 Proposed approach
343(3)
14.4 Experiment and results
346(2)
14.4.1 Dataset
346(1)
14.4.2 Performance analysis
346(2)
14.5 Conclusions and future scope
348(3)
References
349(2)
15 On multi-class aerial image classification using learning machines
351(34)
Qurban A. Memon
Najiya Valappil
15.1 Introduction
352(1)
15.2 Learning approaches
352(8)
15.2.1 Deep learning networks
353(4)
15.2.2 Feature learning
357(1)
15.2.3 Challenges for deep learning
358(1)
15.2.4 Challenges related to aerial video classification
359(1)
15.2.5 Applications
359(1)
15.3 Learning architecture and classification
360(6)
15.3.1 Supervised learning architectures
360(1)
15.3.2 Unsupervised learning
361(1)
15.3.3 Deep learning for planning and situational awareness
362(1)
15.3.4 Deep learning for motion control
362(1)
15.3.5 Object detection
363(1)
15.3.6 Classification
364(2)
15.4 Training
366(4)
15.4.1 Weight initialization
366(1)
15.4.2 Convolutional methods
367(1)
15.4.3 Activation functions
367(1)
15.4.4 Subsampling or pooling layer
368(1)
15.4.5 Optimization techniques
368(2)
15.4.6 Benchmark datasets
370(1)
15.5 Energy efficiency in learning approaches
370(1)
15.6 Performance metrics
370(4)
15.7 Development kits and frameworks
374(1)
15.8 Discussions and future directions
375(10)
References
376(9)
16 Machine learning methodology toward identification of mature citrus fruits
385(54)
Veena Nayak
Sushma P. Holla
Akshaya Kumar
C. Gururaj
16.1 Introduction
385(3)
16.1.1 Harvesting
386(1)
16.1.2 Farm automation
386(1)
16.1.3 Fruit detection
386(1)
16.1.4 Proposed method
387(1)
16.2 Literature survey
388(2)
16.3 Implementation
390(26)
16.3.1 Image acquisition and data collection
392(1)
16.3.2 Pre-processing
392(8)
16.3.3 Feature extraction
400(9)
16.3.4 Machine learning model and database formation
409(2)
16.3.5 Data retrieval
411(1)
16.3.6 Match with dataset or testing the model
411(1)
16.3.7 Display result
412(1)
16.3.8 Application design
413(3)
16.4 Experiments and result
416(18)
16.4.1 Qualification measures
416(1)
16.4.2 Training result
417(4)
16.4.3 Testing result
421(13)
16.5 Conclusion
434(5)
16.5.1 Future scope
435(1)
References
435(4)
17 Automated detection of defects and grading of cashew kernels using machine learning
439(28)
S. V. Veenadevi
C. Srinivasan Padmavathi
17.1 Introduction
440(2)
17.1.1 Related work
441(1)
17.1.2 Proposed methodology
442(1)
17.2 Defects and grades of cashew kernels
442(3)
17.2.1 Cashew kernel manufacturing process
443(1)
17.2.2 Defects of cashew kernel
443(1)
17.2.3 Grades of cashew kernel
444(1)
17.3 Implementation of the methodology
445(9)
17.3.1 Image preprocessing and segmentation
447(3)
17.3.2 Feature extraction
450(3)
17.3.3 Classification
453(1)
17.4 Results and discussions
454(9)
17.5 Conclusion
463(4)
References
463(4)
Index 467
Chiranji Lal Chowdhary is an associate professor in the School of Information Technology and Engineering at VIT University, where he has been since 2010. He received a B.E. (CSE) from MBM Engineering College at Jodhpur in 2001 and M.Tech. (CSE) from the M.S. Ramaiah Institute of Technology at Bangalore in 2008. He received his Ph.D. in Information Technology and Engineering from the VIT University Vellore in 2017. From 2006 to 2010, he worked at M.S. Ramaiah Institute of Technology in Bangalore, eventually as a Lecturer. His research interests span both computer vision and image processing. Much of his work has been on images, mainly through the application of image processing, computer vision, pattern recognition, machine learning, biometric systems, deep learning, soft computing, and computational intelligence. He has given few invited talks on medical image processing. He is the editor/co-editor of five books and is the author of over 40 articles on computer science. He filed two patents deriving from his research.



Mamoun Alazab is an associate professor at the College of Engineering, IT, and Environment, Charles Darwin University, Australia. His multidisciplinary research in cyber security and digital forensics focuses on cybercrime detection and prevention including cyber terrorism and cyber warfare. He works closely with government and industry on projects including IBM, the Australian Federal Police (AFP), the Australian Communications and Media Authority (ACMA), the United Nations Office on Drugs and Crime (UNODC), and the Attorney General's Department. He is a Senior Member of the IEEE and founding chair of the IEEE Northern Territory (NT) Subsection. He holds a Ph.D. degree in Computer Science from the School of Science, Information Technology, and Engineering, Federation University of Australia.



Ankit Chaudhary is an assistant professor at the Department of Computer Science, University of Missouri at Saint Louis, USA. His research focuses on data science, computer vision and cyber security. He has authored three books. He is an associate editor and on the editorial board of several International Journals. He is a member of the IEEE. He received his Ph.D. degree in Computer Engineering from CSIR-CEERI, India.



Saqib Hakak is an assistant professor at the Canadian Institute for Cybersecurity, the University of New Brunswick, Fredericton, Canada. His current research interests include fake news detection, security and privacy, anomaly detection, natural language processing, and applications of AI. He has worked on numerous industrial projects involving IBM Canada, and TD Bank, Bell Canada. He received a complimentary ACM professional membership based on his services to the research community. He holds a Ph.D. degree from the Faculty of Computer Science and Information Technology, University of Malaya, Malaysia.



Thippa Reddy Gadekallu is an associate professor at the School of Information Technology and Engineering, VIT, Vellore, India. His areas of research include machine learning, deep neural networks, internet of things, and blockchain. He holds a Ph.D. degree in Data Mining from VIT, India.
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97818395332426e.html
Märksõnad: