Hybrid Intelligence for Image Analysis and Understanding E-Book
108,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
A synergy of techniques on hybrid intelligence for real-life image analysis Hybrid Intelligence for Image Analysis and Understanding brings together research on the latest results and progress in the development of hybrid intelligent techniques for faithful image analysis and understanding. As such, the focus is on the methods of computational intelligence, with an emphasis on hybrid intelligent methods applied to image analysis and understanding. The book offers a diverse range of hybrid intelligence techniques under the umbrellas of image thresholding, image segmentation, image analysis and video analysis. Key features: * Provides in-depth analysis of hybrid intelligent paradigms. * Divided into self-contained chapters. * Provides ample case studies, illustrations and photographs of real-life examples to illustrate findings and applications of different hybrid intelligent paradigms. * Offers new solutions to recent problems in computer science, specifically in the application of hybrid intelligent techniques for image analysis and understanding, using well-known contemporary algorithms. The book is essential reading for lecturers, researchers and graduate students in electrical engineering and computer science.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 758
Veröffentlichungsjahr: 2017
Ähnliche
Table of Contents
Cover
Title Page
Copyright
Dedication
Editor Biographies
List of Contributors
Foreword
Preface
About the Companion website
Chapter 1: Multilevel Image Segmentation Using Modified Genetic Algorithm (MfGA)-based Fuzzy C-Means
1.1 Introduction
1.2 Fuzzy
C
-Means Algorithm
1.3 Modified Genetic Algorithms
1.4 Quality Evaluation Metrics for Image Segmentation
1.5 MfGA-Based FCM Algorithm
1.6 Experimental Results and Discussion
1.7 Conclusion
References
Chapter 2: Character Recognition Using Entropy-Based Fuzzy C-Means Clustering
2.1 Introduction
2.2 Tools and Techniques Used
2.3 Methodology
2.4 Results and Discussion
2.5 Conclusion and Future Scope of Work
References
Appendix
Chapter 3: A Two-Stage Approach to Handwritten Indic Script Identification
3.1 Introduction
3.2 Review of Related Work
3.3 Properties of Scripts Used in the Present Work
3.4 Proposed Work
3.5 Experimental Results and Discussion
3.6 Conclusion
Acknowledgments
References
Chapter 4: Feature Extraction and Segmentation Techniques in a Static Hand Gesture Recognition System
4.1 Introduction
4.2 Segmentation Techniques
4.3 Feature Extraction Techniques
4.4 State of the Art of Static Hand Gesture Recognition Techniques
4.5 Results and Discussion
4.6 Conclusion
Acknowledgment
References
Chapter 5: SVM Combination for an Enhanced Prediction of Writers' Soft Biometrics
5.1 Introduction
5.2 Soft Biometrics and Handwriting Over Time
5.3 Soft Biometrics Prediction System
5.4 Experimental Evaluation
5.5 Discussion and Performance Comparison
5.6 Conclusion
References
Chapter 6: Brain-Inspired Machine Intelligence for Image Analysis: Convolutional Neural Networks
6.1 Introduction
6.2 Convolutional Neural Networks
6.3 Toward Understanding the Brain, CNNs, and Images
6.4 Conclusion
References
Chapter 7: Human Behavioral Analysis Using Evolutionary Algorithms and Deep Learning
7.1 Introduction
7.2 Human Action Recognition Using Evolutionary Algorithms and Deep Learning
7.3 Experimental Study
7.4 Conclusions and Future Work
References
Chapter 8: Feature-Based Robust Description and Monocular Detection: An Application to Vehicle tracking
8.1 Introduction
8.2 Extraction of Local Features by SIFT and SURF
8.3 Global Features: Real-Time Detection and Vehicle Tracking
8.4 Vehicle Detection and Validation
8.5 Experimental Study
8.6 Conclusions
References
Chapter 9: A GIS Anchored Technique for Social Utility Hotspot Detection
9.1 Introduction
9.2 The Technique
9.3 Case Study
9.4 Implementation and Results
9.5 Analysis and Comparisons
9.6 Conclusions
Acknowledgments
References
Chapter 10: Hyperspectral Data Processing: Spectral Unmixing, Classification, and Target Identification
10.1 Introduction
10.2 Background and Hyperspectral Imaging System
10.3 Overview of Hyperspectral Image Processing
10.4 Spectral Unmixing
10.5 Classification
10.6 Target Detection
10.7 Conclusions
References
Chapter 11: A Hybrid Approach for Band Selection of Hyperspectral Images
11.1 Introduction
11.2 Relevant Concept Revisit
11.3 Proposed Algorithm
11.4 Experiment and Result
11.5 Conclusion
References
Chapter 12: Uncertainty-Based Clustering Algorithms for Medical Image Analysis
12.1 Introduction
12.2 Uncertainty-Based Clustering Algorithms
12.3 Image Processing
12.4 Medical Image Analysis with Uncertainty-Based Clustering Algorithms
12.5 Conclusions
References
Chapter 13: An Optimized Breast Cancer Diagnosis System Using a Cuckoo Search Algorithm and Support Vector Machine Classifier
13.1 Introduction
13.2 Technical Background
13.3 Proposed Breast Cancer Diagnosis System
13.4 Results and Discussions
13.5 Conclusion
13.6 Future Work
References
Chapter 14: Analysis of Hand Vein Images Using Hybrid Techniques
14.1 Introduction
14.2 Analysis of Vein Images in the Spatial Domain
14.3 Analysis of Vein Images in the Frequency Domain
14.4 Comparative Analysis of Spatial and Frequency Domain Systems
14.5 Conclusion
References
Chapter 15: Identification of Abnormal Masses in Digital Mammogram Using Statistical Decision Making
15.1 Introduction
15.2 Previous Works
15.3 Proposed Method
15.4 Experimental Result
15.5 Result Evaluation
15.6 Comparative Analysis
15.7 Conclusion
Acknowledgments
References
Chapter 16: Automatic Detection of Coronary Artery Stenosis Using Bayesian Classification and Gaussian Filters Based on Differential Evolution
16.1 Introduction
16.2 Background
16.3 Proposed Method
16.4 Computational Experiments
16.5 Concluding Remarks
Acknowledgment
References
Chapter 17: Evaluating the Efficacy of Multi-resolution Texture Features for Prediction of Breast Density Using Mammographic Images
17.1 Introduction
17.2 Materials and Methods
17.3 Results
17.4 Conclusion and Future Scope
References
Index
End User License Agreement
Pages
xvii
xviii
xix
xx
xxi
xxii
xxiii
xxiv
xxv
xxvii
xxviii
xxix
xxx
xxxi
xxxii
xxxiii
xxxv
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
153
154
155
156
157
158
159
160
161
162
163
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
283
284
285
286
287
288
289
290
291
292
293
294
295
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
Guide
Cover
Table of Contents
Foreword
Preface
Begin Reading
List of Illustrations
Chapter 1: Multilevel Image Segmentation Using Modified Genetic Algorithm (MfGA)-based Fuzzy C-Means
Figure 1.1 Flowchart of MfGA-based FCM algorithm.
Figure 1.2 8-class segmented 256×256 grayscale Lena image with the class levels obtained by (a–d) FCM, (e–h) GA-based FCM, and (i–l) MfGA-based FCM algorithms of four results of Table 1.1, with as the quality measure.
Figure 1.3 8-class segmented 256×256 grayscale Lena image with the class levels obtained by (a–d) FCM, (e–h) GA-based FCM, and (i–l) MfGA-based FCM algorithms of four results of Table 1.2, with as the quality measure.
Figure 1.4 8-class segmented 256×256 grayscale peppers image with the class levels obtained by (a–d) FCM, (e–h) GA-based FCM, and (i–l) MfGA-based FCM algorithms of four results of Table 1.4, with as the quality measure.
Figure 1.5 8-class segmented 256×256 grayscale peppers image with the class levels obtained by (a–d) FCM, (e–h) GA-based FCM, and (i–l) MfGA-based FCM algorithms of four results of Table 1.5, with as the quality measure.
Figure 1.6 8-class segmented 256×256 grayscale baboon image with the class levels obtained by (a–d) FCM, (e–h) GA-based FCM, and (i–l) MfGA-based FCM algorithms of four results of Table 1.7, with as the quality measure.
Figure 1.7 8-class segmented 256×256 grayscale baboon image with the class levels obtained by (a–d) FCM, (e–h) GA-based FCM, and (i–l) MfGA-based FCM algorithms of four results of Table 1.8, with as the quality measure.
Chapter 2: Character Recognition Using Entropy-Based Fuzzy C-Means Clustering
Figure 2.1 Letters used for training the algorithm: (a) Arial, (b) Bardely, (c) Calibri, (d) Cambria, and (e) Times New Roman.
Figure 2.2 (a) Raw image, (b) converted binary image, and (c) boundary of the image.
Figure 2.3 Flow diagram of preprocessing technique.
Figure 2.4 Longest run feature row-wise and column-wise for a image.
Figure 2.5 Flowchart of the proposed methodology for classification and recognition.
Figure 2.6 Visualization of cluster centers of input data obtained using the (a) FCM algorithm, (b) EFC algorithm, and (c) EFCM algorithm.
Chapter 3: A Two-Stage Approach to Handwritten Indic Script Identification
Figure 3.1 Segments of sample document pages written in: (a) Bangla, (b) Devanagari, (c) Gurumukhi, (d) Oriya, (e) Malayalam, (f) Telugu, (g) Urdu, and (h) Roman scripts, respectively.
Figure 3.2 Wavelet decomposition tree for a 2D image using both low-pass and high-pass filters.
Figure 3.3 Pictorial description of the LL, HL, LH, and HH components after applying Haar wavelet transform to the original grayscale word image written in Bangla script.
Figure 3.4 Illustration of a single projection at a specified rotation angle on a handwritten Bangla word image.
Figure 3.5 Horizontal and vertical projections of a word image written in Bangla script.
Figure 3.6 Pictorial description of the geometry of the radon transformation.
Figure 3.7 Illustration of RT of the word images written in (a) Bangla, (b) Devanagari, (c) Gurumukhi, (d) Oriya, (e) Malayalam, (f) Telugu, (g) Urdu, and (h) Roman scripts, respectively.
Figure 3.8 Graph for: (a) normal distribution, (b) positively skewed distribution, and (c) negatively skewed distribution.
Figure 3.9 Illustration of (a) mesokurtic, (b) leptourtic, and (c) platykurtic curves.
Figure 3.10 Pictorial representation of the two-stage approach of the proposed script identification technique.
Figure 3.11 Sample word images of eight handwritten scripts from our database written in: (a) Bangla, (b) Devanagari, (c) Gurumukhi, (d) Oriya, (e) Malayalam, (f) Telugu, (g) Urdu, and (h) Roman, respectively.
Figure 3.12 Graphical comparison of model building time (seconds) required by seven different classifiers.
Figure 3.13 Comparison of multiple classifiers for: (a) Nemenyi's test and (b) Bonferroni–Dunn's test.
Figure 3.14 Confusion matrix for a classification rule.
Figure 3.15 Graph showing the performance of SVM classifier on the ROC curve for eight handwritten scripts.
Figure 3.16 Samples of successfully classified handwritten word images written in: (a) Bangla, (b) Devanagari, (c) Gurumukhi, (d) Oriya, (e) Malayalam, (f) Telugu, (g) Urdu, and (h) Roman scripts, respectively.
Figure 3.17 Sample handwritten word images misclassified by the present technique due to the presence of: (a) a significantly smaller number of characters constituting the word; (b) skewness; (c–e) structural similarity in Devanagari (misclassified as Gurumukhi), Gurumukhi (misclassified as Devanagari), and Oriya (misclassified as Bangla); (f) Matra-like structure in Roman script; and (g–h) abrupt spaces in Malayalam and Telugu scripts, respectively.
Chapter 4: Feature Extraction and Segmentation Techniques in a Static Hand Gesture Recognition System
Figure 4.1 Operational flowchart of proposed static hand gesture recognition system.
Figure 4.2 Flowchart of hand region extraction method.
Figure 4.3 Otsu method for thresholding:(a) gray-level image, (b) bimodal histogram, and (c) segmented binary image.
Figure 4.4 Hand gesture images captured in different angles.
Figure 4.5 Block diagram of homomorphic filtering.
Figure 4.6 (a) RGB color image, (b) segmented image, and (c) ROI after morphological operation.
Figure 4.7 Basic steps of feature extraction.
Figure 4.8 Zoning topology.
Figure 4.9 Uniform Background Database: (a) digit 1, (b) digit 2, and (c) digit 3.
Figure 4.10 Complex Background Database: (a) digit 0, (b) digit 1, and (c) digit 2.
Figure 4.11 (a) Input gesture, (b) YCbCr segmented image, and (c) extracted hand region after morphological operation.
Figure 4.12 (a) input gesture, (b) segmented image by our proposed method, and (c) extracted hand region after morphological operation.
Figure 4.13 ASL similar-shape gestures: (a) 7, (b) 8, and (c) 9.
Chapter 5: SVM Combination for an Enhanced Prediction of Writers' Soft Biometrics
Figure 5.1 Soft-biometrics prediction system.
Figure 5.2 HOG feature calculated on a handwritten text image.
Figure 5.3 GLBP feature extraction for each pixel.
Figure 5.4 IAM data set samples.
Figure 5.5 KHATT data set samples.
Figure 5.6 LBP operators; performances for gender prediction on the IAM-1 corpus.
Figure 5.7 Influence of the grid size for gender prediction on the IAM-1 corpus.
Chapter 6: Brain-Inspired Machine Intelligence for Image Analysis: Convolutional Neural Networks
Figure 6.1 A neuron.
Figure 6.2 Hidden layers.
Figure 6.3 A perceptron.
Figure 6.4 Gradient of a single weight.
Figure 6.5 Example of applying convolution to images.
Figure 6.6 Convolution kernels.
Figure 6.7 Edge detection in images.
Figure 6.8 Sobel operator.
Figure 6.9 Convolutional neural network: architecture.
Figure 6.10 Connections in a regular neural network versus a convolutional neural network.
Figure 6.11 Example of parameter sharing among neurons.
Figure 6.12 Example of pooling.
Figure 6.13 Input image.
Figure 6.14 Output from the first sublayer of the first layer.
Figure 6.15 Output from the first layer.
Figure 6.16 Output from the second layer.
Figure 6.18 Output from the fourth layer.
Figure 6.19 Output from the fifth layer.
Chapter 7: Human Behavioral Analysis Using Evolutionary Algorithms and Deep Learning
Figure 7.1 Pictorial representation of various steps in the proposed classification system. Best viewed in color.
Figure 7.2 Action bank features of observations in KTH: (i–iii) are videos of boxing action and (iv–vi) are videos of running.
Figure 7.3 CNN classifier for recognizing actions in videos.
Figure 7.4 Analysis of classification errors of solutions generated by the proposed hybrid training approach.
Figure 7.5 Visualization of EA population generated by the proposed approach for
Set-1
of UCF50: (i) after initialization by EA and (ii) after training the solutions with BPA for epochs.
Figure 7.6 Visualization of EA population generated by the proposed approach for
Set-2
to
Set-5
of UCF50. The sub-Figure (i), (ii) correspond to
Set-2
; (iii), (iv) are for
Set-3
; (v), (vi) correspond to
Set-4
and (vii), (viii) are for
Set-5
.
Figure 7.7 Visualization of EA population generated for KTH: (i) visualization after initialization by EA and (ii) visualization after training with BPA.
Figure 7.8 Visualization of chromosomes (candidate solutions) generated by the proposed hybrid training approach for UCF50: (a) The pictorial representation of convolution masks and seed values corresponding to a chromosome
X
of size 64; (b) chromosomes generated for the KTH data set; and (c–g) correspond to chromosomes generated for UCF50
Set-1
to
Set-5
, respectively.
Chapter 8: Feature-Based Robust Description and Monocular Detection: An Application to Vehicle tracking
Figure 8.1 Illustration of Lowe's matching method.
Figure 8.2 Generic points aided robust description (GPRD) matching scheme.
Figure 8.3 Matching with Lowe's SIFT description.
Figure 8.4 Matching with GPRD descriptors.
Figure 8.5 Visual comparison of GPRD and Lowe's SIFT description: The frame on the left side (blue colored) is the result of GPRD, and the frame on the right side (red colored) is the result of the SIFT description.
Figure 8.6 Assessment of Haar-like features in the context of face recognition.
Figure 8.7 Recognition and tracking scheme.
Figure 8.8 Scan for lines' vertical response (right), and discontinuity on lines (left).
Figure 8.9 Screenshots of the recognition and tracking results on the Istanbul TEM highway.
Figure 8.10 The responses of true matchings for SIFT, SURF, and SIFT GPRD–based detection.
Figure 8.11 The responses of false matchings.
Figure 8.12 Evaluation of SIFT GPRD in a road traffic video.
Figure 8.13 Screenshots of recognition and tracking system result for TEM highway during rush hour.
Figure 8.14 Screenshots of recognition and tracking system results for LISA-Q Front FOV 1 during rush hour.
Figure 8.15 Comparison of true positive per frame metrics for global and local features.
Figure 8.16 Comparison of false positive per frame metrics for global and local features.
Chapter 9: A GIS Anchored Technique for Social Utility Hotspot Detection
Figure 9.1 Map of Egra police station, West Bengal, India.
Figure 9.2 Existing police stations of Egra.
Figure 9.3 Cluster formation after acquiring ranks.
Figure 9.4 Rank indicator.
Figure 9.5 Hotspot zone formation after considering Condition(1) and Condition(2).
Figure 9.6 Depicting the encircled “red” hotspot zone.
Figure 9.7 Splitting the “red” zone into two clusters, and .
Figure 9.8 Depicting suitable locations for construction of beat houses.
Figure 9.9 Before.
Figure 9.10 After.
Figure 9.11 Choosing options.
Figure 9.12 Creating a new profile.
Figure 9.13 Opening an existing profile.
Figure 9.14 Digitization of a raster map.
Figure 9.15 Data association.
Figure 9.16 “Click Here” button.
Figure 9.17 Crime-Info window.
Figure 9.18 Number of cases received against respective crime types.
Figure 9.19 Buttons under “Police Station” label.
Figure 9.20 In the year 2011 (3 suggested police stations).
Figure 9.21 In the year 2012 (2 suggested police stations).
Figure 9.22 In the year 2013 (2 suggested police stations).
Chapter 10: Hyperspectral Data Processing: Spectral Unmixing, Classification, and Target Identification
Figure 10.1 Hyperspectral imaging concept.
Figure 10.2 Diagrammatic representation of the schematics for a commonly used hyperspectral imaging system.
Figure 10.3 General processing steps involved for hyperspectral data.
Figure 10.4 Hyperspectral unmixing processing chain.
Figure 10.5 Conceptual diagram of a simple linear mixture model geometry.
Figure 10.6 Nonlinear mixture model.
Figure 10.7 Double concentric sliding window.
Chapter 11: A Hybrid Approach for Band Selection of Hyperspectral Images
Figure 11.1 Electromagnetic spectrum [5].
Figure 11.2 AVIRIS hyperspectral image cube [6].
Figure 11.3 Basic biological immune system [33].
Figure 11.4 Flowchart of the clonal selection technique.
Figure 11.5 Flowchart of the proposed technique.
Figure 11.6 Indiana Pines image. (a) Indiana Pines image. (b) Ground truth.
Figure 11.7 Pavia image. (a) Three-band color composite image. (b) Ground truth.
Figure 11.8 Overall accuracy of the proposed algorithm on the Indiana data set.
Figure 11.9 Comparison of the proposed method with MI, WaLuDi, and TMI in terms of overall accuracy for the Indiana data set.
Figure 11.10 Comparison of the proposed method with MI, WaLuDi, and TMI in terms of overall accuracy for the Pavia Center data set.
Figure 11.11 Basics of a SVM.
Figure 11.12 Comparison of the proposed method (2D PCA and fuzzy KNN) with 2D PCA and SVM in terms of overall accuracy for the Indiana data set.
Figure 11.13 Comparison of the proposed method (2D PCA) and LDA in terms of overall accuracy for the Indiana data set.
Chapter 12: Uncertainty-Based Clustering Algorithms for Medical Image Analysis
Figure 12.1 MRI image segmentation using (a) FCM, (b) , and (c) .
Figure 12.2 Segmentation results on (a) original image, (b) same image with mixed noise, results of (c) FCM_S1, (d) FCM_S2, (e) EnFCM, (f) FGFCM_S1, (g) FGFCM_S2, and (h) FGFCM.
Figure 12.3 MRI image – speckle noise.
Figure 12.4 Noisy image segmentation. (a) FCM, (b) sFCM1,1, (c) , (d) , (e) , (f) sIFCM, and (g) .
Figure 12.5 (a) Original image. Segmented images of leukemia using (b) FCM, (c) , (d) , and (e) .
Figure 12.6 (a) Original image. Segmented images of leukemia using (b) FCM, (c) IFCM, (d) , (e) , and (f) .
Chapter 13: An Optimized Breast Cancer Diagnosis System Using a Cuckoo Search Algorithm and Support Vector Machine Classifier
Figure 13.1 Proposed breast cancer diagnosis system.
Figure 13.2 Sample mammographic images.
Figure 13.3 Enhanced mammographic images.
Figure 13.4 Otsu's thresholded and morphological segmented breast tissues.
Chapter 14: Analysis of Hand Vein Images Using Hybrid Techniques
Figure 14.1 The basic block diagram of a multimodal biometric recognition system.
Figure 14.2 Flow diagram of the analysis of vein images in the spatial domain.
Figure 14.3 Input hand vein images (first row). Preprocessed vein images (second row).
Figure 14.4 Output of Gabor filter for Palm vein images with : (a) , ; (b) , ; (c) , ; (d) , ; (e) , ; (f) , ; (g) , ; (h) , ; and (i) , .
Figure 14.6 Output of Gabor filter for Wrist vein images with : (a) , ; (b) , ; (c) , ; (d) , ; (e) , ; (f) , ; (g) , ; (h) , ; and (i) , .
Figure 14.7 Output of Gabor filter for Finger vein images with : (a) , ; (b) , ; (c) , ; (d) , ; (e) , ; (f) , ; (g) , ; (h) , ; and (i) , .
Figure 14.8 Fusion of two modalities of hand vein features.
Figure 14.9 Fusion of three modalities of hand vein features.
Figure 14.10 Fusion of all four modalities of hand vein features.
Figure 14.11 Flow diagram of the analysis of vein images in the frequency domain.
Figure 14.12 Fusion for the hand vein images in the frequency domain.
Figure 14.13 A linear support vector machine.
Figure 14.14 Preprocessed hand vein images at various stages.
Figure 14.15 Contourlet-transformed vein images: (a) dorsal hand vein, (b) palm vein, (c) wrist vein, and (d) finger vein.
Chapter 15: Identification of Abnormal Masses in Digital Mammogram Using Statistical Decision Making
Figure 15.1 Schematic diagram of the proposed method.
Figure 15.2 (a) The original mammogram image and (b) the prepared mammogram image (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society. [31]
Figure 15.3 Full and complete binary tree.
Figure 15.4 Enhanced mammogram image (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.5 (a) Edge map of Level 1, (b) edge map of Level 2, and (c) edge map of Level 3 (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.6 (a) The edge map of a mammogram, (b) showing the layers of pectoral muscle, and (c) showing inverted triangles marked by different gray shades (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.7 Pectoral muscle lies within gray-shaded derived rectangular area (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.8 Isolated pectoral boundary (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.9 The technique of traversing process. (a) Black cell represents the current pixel, gray cell is representative of already traversed pixel, and the rest are the path for further traversing. (b) The priority of selection of neighbor is clockwise.
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.10 Detected breast contour (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.11 (a) Breast ROI and (b) Boundaries of anatomical regions within breast ROI (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.12 Intensity distribution of regions after coloring (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.13 (a) Highlighted regions with abnormal masses and (b) boundary of abnormal regions (MIAS mdb184.L).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.14 MIAS mdb272.L: (a) mammogram image, (b) segmented anatomical regions without highlighted abnormality, and (c) derived image showing absence of boundary of abnormal region(s).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.15 MIAS mdb028.L: (a) mammogram image, (b) segmented anatomical regions with highlighted abnormality, and (c) derived image showing boundary of abnormal region(s).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.16 MIAS mdb001.R: (a) mammogram image, (b) segmented anatomical regions with highlighted abnormality, and (c) derived image showing boundary of abnormal region(s).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.17 MIAS mdb145.R: (a) mammogram image, (b) segmented anatomical regions with highlighted abnormality, and (c) derived image showing boundary of abnormal region(s).
Source
: Suckling (1994) [31]. Reproduced with permission of Mammographic Image Analysis Society.
Figure 15.18 The
Z
score analysis graph for mammogram mdb272.L.
Figure 15.19 score analysis graph for mammogram mdb028.L.
Figure 15.20 The
Z
score analysis graph for mammogram mdb001.R.
Figure 15.21 The
Z
score analysis graph for mammogram mdb145.R.
Figure 15.22 Empirical ROC curve for tumor identification.
Chapter 16: Automatic Detection of Coronary Artery Stenosis Using Bayesian Classification and Gaussian Filters Based on Differential Evolution
Figure 16.1 Two X-ray angiograms with detection of coronary stenosis performed by a cardiologist.
Figure 16.2 (a) X-ray coronary angiogram. (b) Ground-truth image of angiogram in (a). (c) Gaussian profile of the method of Chaudhuri
et al.
[16]. Second row: Template using , , , with , and , respectively, and resulting filtered image in (f). Last row: Matching template using , , , with , and , respectively, and resulting filtered image in (i).
Figure 16.3 Ackley function in two dimensions. (a) Isometric view in and , and (b) level plot of the function, where the optimal value is located at and .
Figure 16.4 Numerical example for solving the 2D Ackley function using DE as an optimization strategy.
Figure 16.5 First row: Segmentation results using the Ridler and Calvard method. The remaining three rows illustrate the results of length filtering using 100, 200, and 500 pixels as connected components, respectively.
Figure 16.6 (a) X-ray coronary angiogram. (b) Skeleton of segmented vessel. (c) Addition of skeleton and boundary pixels. (d) Skeleton using normalized intensities as Euclidean distance. (e) Separation of vessel segments using bifurcation pixels. (f, g) Detection of local minima points over Gaussian filter response and original angiogram, respectively. (h) Stenosis detection marked in a black circle by cardiologist.
Figure 16.7 (a) Stenosis pattern of pixels. (b) Histogram of vessel width estimation of pattern in (a). (c) No stenosis pattern of pixels. (d) Histogram of vessel width estimation of pattern in (c).
Figure 16.8 First row: Subset of X-ray angiograms. Second row: Ground-truth images. The remaining six rows present the Gaussian filter response of the methods of Kang
et al.
[25], Al-Rawi
et al.
[21], Cruz
et al.
[27], Chaudhuri
et al.
[16], and Cinsdikici
et al.
[20], and the proposed method, respectively.
Figure 16.9 First row: Subset of X-ray angiograms. Second row: Ground-truth images. The remaining five rows present the segmentation results of the methods of Kapur
et al.
[33], histogram concavity [34], Pal and Pal [35], RATS [36], and Ridler and Calvard [32], respectively.
Figure 16.10 First column: Subset of X-ray angiograms. Second column: Ground-truth images. Third column: Segmentation result obtained from the proposed method. Last column: Product between segmentation result and input angiogram.
Figure 16.11 First row: Subset of patterns of no-stenosis cases. Second column: Subset of vessel stenosis patterns.
Chapter 17: Evaluating the Efficacy of Multi-resolution Texture Features for Prediction of Breast Density Using Mammographic Images
Figure 17.1 CAD system for breast density classification.
Note
: Shaded blocks indicate the steps involved in present work.
Figure 17.2 MIAS database breast density classification.
Figure 17.3 Sample mammograms showing typical cases of breast tissue density: (a) typical F tissue , (b) typical FG tissue , and (c) typical DG tissue .
Figure 17.4 Sample mammograms showing atypical cases of breast tissue density: (a) atypical F tissue , (b) atypical FG tissue , and (c) atypical DG tissue .
Figure 17.5 Database description.
Figure 17.6 ROI extraction protocol .
Figure 17.7 Sample ROIs: (a) typical F ROI ; (b) typical FG ROI ; (c) typical DG ROI ; (d) atypical F ROI ; (e) atypical FG ROI ; and (f) atypical DG ROI .
Figure 17.8 Block diagram: workflow for prediction of breast density.
Figure 17.9 Different feature extraction techniques used in texture analysis. Note: GLCM: gray-level co-occurence matrix; GLRLM: gray-level run length matrix; NGTDM: neighborhood gray tone difference matrix; SFM: statistical feature matrix; FPS: Fourier power spectrum; STFT: short-time Fourier transform; 2D-DWT: two-dimensional discrete wavelet transform; WPT: wavelet packet transform; NSCT: non-subsampled countourlet transform; NSST: non-subsampled shearlet transform.
Figure 17.10 Time-domain, frequency-domain, STFT, and wavelet analysis of a signal.
Figure 17.11 Process of wavelet analysis of an image.
Figure 17.12 Wavelet transform of sample Lena image.
Figure 17.13 Wavelet transform of image F
mdb132
.
Figure 17.14 Wavelet decomposition of an image up to the second level.
Figure 17.15 (a) 2D wavelet decomposition of image up to the second level. (b) 2D wavelet decomposition of sample image using a Haar wavelet filter up to the second level.
Figure 17.16 (a) Haar wavelet function and (b) Haar scaling function .
Figure 17.17 SVM classifier for linearly separable data.
Figure 17.18 Example to illustrate the SVM algorithm.
Figure 17.19 SVM for nonlinearly separable data.
Figure 17.20 Flowchart of proposed CAD system design.
Figure 17.21 Proposed CAD system design.
List of Tables
Chapter 1: Multilevel Image Segmentation Using Modified Genetic Algorithm (MfGA)-based Fuzzy C-Means
Table 1.1 Class boundaries and evaluated segmentation quality measures by different algorithms for different classes of the Lena image
Table 1.2 Class boundaries and evaluated segmentation quality measures by different algorithms for different classes of the Lena image
Table 1.3 Different algorithm-based means and standard deviations using different types of fitness functions and mean of time taken by different algorithms for the Lena image
Table 1.4 Class boundaries and evaluated segmentation quality measures by different algorithms for different classes of the peppers image
Table 1.5 Class boundaries and evaluated segmentation quality measures by different algorithms for different classes of the peppers image
Table 1.6 Different algorithm-based mean and standard deviation using different types of fitness functions and mean of time taken by different algorithms for the peppers image
Table 1.7 Class boundaries and evaluated segmentation quality measures by different algorithms for different classes of the baboon image
Table 1.8 Class boundaries and evaluated segmentation quality measures by different algorithms for different classes of the baboon image
Table 1.9 Different algorithm-based mean and standard deviation using different types of fitness functions and mean of time taken by different algorithms for the baboon image
Chapter 2: Character Recognition Using Entropy-Based Fuzzy C-Means Clustering
Table 2.1 Output of different clustering algorithms for input data
Table 2.2 Results of recognition of alphabets with three algorithms for Times New Roman
Table 2.3 Recognition accuracy with respect to each font
Chapter 3: A Two-Stage Approach to Handwritten Indic Script Identification
Table 3.1 Important information related to scripts [26] used in the present work
Table 3.2 Success rates of the proposed script identification technique using seven well-known classifiers
Table 3.3 Recognition accuracies of seven classifiers and their corresponding ranks on 12 different data sets (ranks in parentheses are used for performing the Friedman test)
Table 3.4 Statistical performance measures along with their respective means (shaded in gray) achieved by the proposed technique for eight handwritten scripts
Table 3.5 Comparison of statistical performance parameters for the four cases (best case is styled in bold)
Table 3.6 Comparison of the present script identification result with state-of-the art methods
Chapter 4: Feature Extraction and Segmentation Techniques in a Static Hand Gesture Recognition System
Table 4.1 User-dependent and user-independent classification results
Table 4.2 Confusion matrix of the Krawtchouk moment for user-independent condition
Table 4.3 Confusion matrix of the Tchebichef moment for the user-independent condition
Table 4.4 Confusion matrix of the geometric moment for the user-independent condition
Table 4.5 Classification result of Krawtchouk moment zonal features
Table 4.6 Classification results of serial, parallel, and MLE-based hidden feature fusion
Table 4.7 Classification result of
F
-ratio-based enhanced features
Chapter 5: SVM Combination for an Enhanced Prediction of Writers' Soft Biometrics
Table 5.1 Influence of SVM kernels for gender prediction on the IAM-1 corpus (%)
Table 5.2 Results of individual systems for gender prediction (%)
Table 5.3 Results of combination systems for gender prediction (%)
Table 5.4 Results of handedness prediction for individual systems (%)
Table 5.5 Results of combination systems for handedness prediction (%)
Table 5.6 Results of handedness prediction for individual systems (%)
Table 5.7 Results of combination systems for age prediction (%)
Table 5.8 State-of-the-art results
Chapter 6: Brain-Inspired Machine Intelligence for Image Analysis: Convolutional Neural Networks
Table 6.1 Conceptual differences between CNN and the brain/visual system
Table 6.2 Activation functions
Table 6.3 Factors and assigned variables for the perceptron model
Chapter 7: Human Behavioral Analysis Using Evolutionary Algorithms and Deep Learning
Table 7.1 The variation in fitness value of EA populations over iterations for UCF50
Table 7.2 Accuracy (in %) of candidate solutions generated for UCF50 using neural network classifier
Table 7.3 Performance (in %) of candidate solutions generated for UCF50 using ELM classifier
Table 7.4 Accuracy (in # misclassified observations) using fusion on UCF50
Table 7.5 Confusion matrix for UCF50
Table 7.6 Performance on the UCF50 data set
Table 7.7 Performance (in %) of CNN features with NN and ELM classifiers generated by the proposed approach [with back-propagation algorithm (BPA) and evolutionary algorithms (EA)] on the KTH data set
Table 7.8 Performance (in %) on the KTH data set
Table 7.9 Accuracy (in %) of CNN classifiers using BPAs, EAs and the hybrid approach for UCF50
Table 7.10 Accuracy (in %) of the proposed approach using NN and ELM classifiers for UCF50
Chapter 8: Feature-Based Robust Description and Monocular Detection: An Application to Vehicle tracking
Table 8.1 Performance results
Table 8.2 Performance results (video data set belonging to [12])
Chapter 9: A GIS Anchored Technique for Social Utility Hotspot Detection
Table 9.1 Hotspot-detecting tasks, depending on respective factors
Table 9.2 Table depicting the least, moderate, and highest crime-prone regions situated in seven police stations throughout West Bengal State, India
Table 9.3 Classification of crime types with respect to their rankings
Table 9.4 Table depicting the numerator and denominator values of respective regions calculated with respect to the crime types (as obtained from Table 9.3)
Table 9.5 Table delineating the final values of 's and 's (data for =1)
Table 9.6 Table delineating the final values of 's and 's (data for =5)
Table 9.7 Table delineating the final values of 's and 's (data for =10)
Table 9.8 Values of the unknown variables used in Equation (9.1)
Table 9.9 The ranks of respective Anchals (arranged in descending order of rank)
Table 9.10 Depicting the comparative study between
K
-means and the proposed method
Table 9.11 Portraying the comparative study between the fuzzy clustering method and proposed method
Table 9.12 Illustrating the comparative study between ISODATA and the proposed method
Table 9.13 The comparative study between STAC and proposed method is drawn out
Table 9.14 Illustrating the advantages of the proposed method over RADIUS methodology
Table 9.15 Comparison with MCE – a technique for hotspot detection related to landslides
Chapter 11: A Hybrid Approach for Band Selection of Hyperspectral Images
Table 11.1 Indian Pines data set: classes with number of samples
Table 11.2 Pavia data set: classes with number of samples
Table 11.3 Selected bands for Indian Pines data set obtained by the proposed method
Table 11.4 Accuracy of classification using the algorithm
Chapter 12: Uncertainty-Based Clustering Algorithms for Medical Image Analysis
Table 12.1 Cluster evaluation results on speckle noise image
Table 12.2 Performance indices of sFCM on leukemia image
Table 12.3 Performance indices of sIFCM on leukemia image
Chapter 13: An Optimized Breast Cancer Diagnosis System Using a Cuckoo Search Algorithm and Support Vector Machine Classifier
Table 13.1 Comparison of classification accuracy of proposed approach with the existing methods
Table 13.2 Comparison of computational costs
Chapter 14: Analysis of Hand Vein Images Using Hybrid Techniques
Table 14.1 Analysis of hand vein images in unimodal mode
Table 14.2 Analysis of hand vein images in multimodal mode
Table 14.3 Frequency domain analysis of hand vein images
Table 14.4 Comparative analysis of related work on vein-based biometric recognition
Chapter 15: Identification of Abnormal Masses in Digital Mammogram Using Statistical Decision Making
Table 15.1 Color lookup table
Table 15.2 Confusion matrix of response data reported from testing
Table 15.3 Observed operating points
Table 15.4 Accuracy measures based on size of mass detected by the proposed method
Table 15.5 Quantitative measures applied to assess the proposed methods
Table 15.6 Comparative analysis of proposed method with others
Chapter 16: Automatic Detection of Coronary Artery Stenosis Using Bayesian Classification and Gaussian Filters Based on Differential Evolution
Table 16.1 Training records for predicting cardiovascular risk
Table 16.2 Comparative analysis of values with the testing set, using the proposed method and five GMF-based methods of the state of the art
Table 16.3 Comparative analysis of five automatic thresholding methods over the Gaussian filter response using the test set of X-ray angiograms
Table 16.4 Results of naive Bayes classifier over the test set of 20 records
Table 16.5 Confusion matrix for the test set of 20 records
Chapter 17: Evaluating the Efficacy of Multi-resolution Texture Features for Prediction of Breast Density Using Mammographic Images
Table 17.1 Description of studies carried out for classification of tissue density as fatty or dense on the MIAS database
Table 17.2 Description of studies carried out for classification of tissue density as fatty, fatty-glandular, dense-glandular, or extremely dense on the MIAS database
Table 17.3 Description of studies carried out for classification of tissue density as fatty, fatty-glandular, or dense-glandular on the MIAS database
Table 17.4 Properties of wavelet filters used
Table 17.5 Description of FDVs
Table 17.6 Experiment descriptions
Table 17.7 Classification performance of SVM classifier using different FDVs
Table 17.8 Classification performance of SSVM classifier with different FDVs
Table 17.9 Comparison of computational time for prediction of testing instances
Hybrid Intelligence for Image Analysis and Understanding
Edited by
Siddhartha Bhattacharyya
RCC Institute of Information Technology India
Indrajit Pan
RCC Institute of Information Technology India
Anirban Mukherjee
RCC Institute of Information Technology India
Paramartha Dutta
Visva-Bharati University India
This edition first published 2017
© 2017 John Wiley & Sons Ltd
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
The right of Siddhartha Bhattacharyya, Indrajit Pan, Anirban Mukherjee, and Paramartha Dutta to be identified as the authors of the editorial material in this work has been asserted in accordance with law.
Registered Office(s)
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats.
Limit of Liability/Disclaimer of Warranty
While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data:
Names: Bhattacharyya, Siddhartha, 1975- editor. | Pan, Indrajit, 1983- editor. | Mukherjee, Anirban, 1972- editor. | Dutta, Paramartha, editor.
Title: Hybrid intelligence for image analysis and understanding / edited by Siddhartha Bhattacharyya, Indrajit Pan, Anirban Mukherjee, Paramartha Dutta.
Description: Hoboken, NJ : John Wiley & Sons, 2017. | Includes index. | Identifiers: LCCN 2017011673 (print) | LCCN 2017027868 (ebook) | ISBN 9781119242932 (pdf) | ISBN 9781119242956 (epub) | ISBN 9781119242925 (cloth)
Subjects: LCSH: Image analysis. | Computational intelligence.
Classification: LCC TA1637 (ebook) | LCC TA1637 .H93 2017 (print) | DDC 621.36/7028563-dc23
LC record available at https://lccn.loc.gov/2017011673
Cover design by Wiley
Cover images: (Background) © Yakobchuk/Gettyimages; (From left to right)
© zmeel/Gettyimages; © tkemot/Shutterstock; © Semnic/Shutterstock;
© Callista Images/Gettyimages; © Karl Ammann/Gettyimages
Dedication
Dedicated to my parents, the late Ajit Kumar Bhattacharyya and the late Hashi Bhattacharyya; my beloved wife, Rashni; my elder sisters, Tamali, Sheuli, and Barnali; my cousin sisters, Sutapa, Mousumi, and Soma; and all my students, who have made this journey enjoyable.
Dr. Siddhartha Bhattacharyya
Dedicated to all my students.
Dr. Indrajit Pan
Dedicated to my respected teachers.
Dr. Anirban Mukherjee
Dedicated to my parents, the late Arun Kanti Dutta and Mrs. Bandana Dutta.
Dr. Paramartha Dutta
Editor Biographies
Dr Siddhartha Bhattacharyya earned his bachelor's in Physics, bachelor's in Optics and Optoelectronics, and master's in Optics and Optoelectronics from University of Calcutta, India, in 1995, 1998, and 2000, respectively. He completed a PhD in computer science and engineering from Jadavpur University, India, in 2008. He is the recipient of the University Gold Medal from the University of Calcutta for his master's in 2012. He is also the recipient of the coveted ADARSH VIDYA SARASWATI RASHTRIYA PURASKAR for excellence in education and research in 2016. He is the recipient of the Distinguished HoD Award and Distinguished Professor Award conferred by Computer Society of India, Mumbai Chapter, India in 2017. He is also the recipient of the coveted Bhartiya Shiksha Ratan Award conferred by Economic Growth Foundation, New Delhi in 2017.
He is currently the Principal of RCC Institute of Information Technology, Kolkata, India. In addition, he is serving as the Dean of Research and Development of the institute from November 2013. Prior to this, he was the Professor and Head of Information Technology of RCC Institute of Information Technology, Kolkata, India, from 2014 to 2017. Before this, he was an Associate Professor of Information Technology in the same institute, from 2011 to 2014. Before that, he served as an Assistant Professor in Computer Science and Information Technology of University Institute of Technology, The University of Burdwan, India, from 2005 to 2011. He was a Lecturer in Information Technology of Kalyani Government Engineering College, India, during 2001–2005. He is a coauthor of four books and the coeditor of eight books, and has more than 175 research publications in international journals and conference proceedings to his credit. He has got a patent on intelligent colorimeter technology. He was the convener of the AICTE-IEEE National Conference on Computing and Communication Systems (CoCoSys-09) in 2009. He was the member of the Young Researchers' Committee of the WSC 2008 Online World Conference on Soft Computing in Industrial Applications. He has been the member of the organizing and technical program committees of several national and international conferences. He served as the Editor-in-Chief of International Journal of Ambient Computing and Intelligence (IJACI) published by IGI Global (Hershey, PA, USA) from July 17, 2014, to November 6, 2014. He was the General Chair of the IEEE International Conference on Computational Intelligence and Communication Networks (ICCICN 2014) organized by the Department of Information Technology, RCC Institute of Information Technology, Kolkata, in association with Machine Intelligence Research Labs, Gwalior, and IEEE Young Professionals, Kolkata Section; it was held at Kolkata, India, in 2014. He is the Associate Editor of International Journal of Pattern Recognition Research. He is the member of the editorial board of International Journal of Engineering, Science and Technology and ACCENTS Transactions on Information Security (ATIS). He is also the member of the editorial advisory board of HETC Journal of Computer Engineering and Applications. He has been the Associate Editor of the International Journal of BioInfo Soft Computing since 2013. He is the Lead Guest Editor of the Special Issue on Hybrid Intelligent Techniques for Image Analysis and Understanding of Applied Soft Computing (Elsevier, Amsterdam). He was the General Chair of the 2015 IEEE International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN 2015) organized by the Department of Information Technology, RCC Institute of Information Technology, Kolkata, in association with IEEE Young Professionals, Kolkata Section and held at Kolkata, India, in 2015. He is the Lead Guest Editor of the Special Issue on Computational Intelligence and Communications in International Journal of Computers and Applications (IJCA) (Taylor & Francis, London) in 2016. He has been the Issue Editor of International Journal of Pattern Recognition Research since January 2016. He was the General Chair of the 2016 International Conference on Wireless Communications, Network Security and Signal Processing (WCNSSP2016) held during June 26–27, 2016, at Chiang Mai, Thailand. He is the member of the editorial board of Applied Soft Computing (Elsevier, Amsterdam).
He has visited several leading universities in several countries like China, Thailand, and Japan for delivering invited lectures. His research interests include soft computing, pattern recognition, multimedia data processing, hybrid intelligence, and quantum computing. Dr Bhattacharyya is a Fellow of Institute of Electronics and Telecommunication Engineers (IETE), India. He is a senior member of Institute of Electrical and Electronics Engineers (IEEE), USA; Association for Computing Machinery (ACM), USA; and International Engineering and Technology Institute (IETI), Hong Kong. He is a member of International Rough Set Society, International Association for Engineers (IAENG), Hong Kong; Computer Science Teachers Association (CSTA), USA; International Association of Academicians, Scholars, Scientists and Engineers (IAASSE), USA; Institution of Engineering and Technology (IET), UK; and Institute of Doctors Engineers and Scientists (IDES), India. He is a life member of Computer Society of India, Optical Society of India, Indian Society for Technical Education, and Center for Education Growth and Research, India.
Dr Indrajit Pan has done his BE in Computer Science and Engineering with Honors from The University of Burdwan in 2005; M.Tech. in Information Technology from Bengal Engineering and Science University, Shibpur, in 2009; and PhD (Engg.) from Indian Institute of Engineering Science and Technology, Shibpur, in 2015. He is the recipient of BESU, University Medal for securing first rank in M.Tech. (IT). He has a couple of national- and international-level research publications and book chapters to his credit. He has attended several international conferences, national-level faculty development programs, workshops, and symposiums.
In this Institute, his primary responsibility is teaching and project guidance at UG (B.Tech.) and PG (M.Tech. and MCA) levels as Assistant Professor of Information Technology (erstwhile Lecturer since joining in February 2006). Apart from this, he has carried out additional responsibility of Single Point of Contact (SPoC) for Infosys Campus Connect Initiative in 2009–2011, and Coordinator of Institute-level UG Project Committee in 2008–2010. At present, his additional responsibility includes Nodal Officer of Institutional Reforms for TEQIP–II Project since 2011 and Member Secretary of Academic Council since 2013. Apart from these, he has actively served as an organizing member of several Faculty Development Programs and International Conferences (ICCICN 2014) for RCCIIT. He has also acted as the Session Chair in an International Conference (ICACCI 2013) and Member of Technical Committee for FICTA 2014. Before joining RCCIIT, Indrajit served Siliguri Institute of Technology, Darjeeling, as a Lecturer in CSE/IT from 2005 to 2006.
Indrajit is a Member of Institute of Electrical and Electronics Engineers (IEEE), USA; and Association for Computing Machinery (ACM), USA.
Dr Anirban Mukherjee did his bachelor's in Civil Engineering in 1994 from Jadavpur University, Kolkata. While in service, he achieved a professional Diploma in Operations Management (PGDOM) in 1998 and completed his PhD on Automatic Diagram Drawing based on Natural Language Text Understanding from Indian Institute of Engineering, Science and Technology (IIEST), Shibpur, in 2014. Serving RCC Institute of Information Technology (RCCIIT), Kolkata, since its inception (in 1999), he is currently an Associate Professor and Head of the Department of Engineering Science & Management at RCCIIT. Before joining RCCIIT, he served as an Engineer in the Scientific & Technical Application Group in erstwhile RCC, Calcutta, for six years. His research interest includes computer graphics, computational intelligence, optimization, and assistive technology. He has coauthored two UG engineering textbooks: one on Computer Graphics and Multimedia and another on Engineering Mechanics. He has also coauthored more than 18 books on Computer Graphics/Multimedia for distance learning courses He holds BBA/MBA/BCA/MCA/B.Sc (Comp.Sc.)/M.Sc (IT) of different universities of India. He has a few international journal articles, book chapters, and conference papers to his credit. He is in the editorial board of International Journal of Ambient Computing and Intelligence (IJACI).
Dr Paramartha Dutta, born 1966, did his bachelor's and master's in Statistics from the Indian Statistical Institute, Calcutta, in the years 1988 and 1990, respectively. He afterward completed his master's degree of technology in Computer Science from the same Institute in 1993, and PhD in engineering from the Bengal Engineering and Science University, Shibpur, in 2005, respectively. He has served in the capacity of research personnel in various projects funded by Government of India, which include DRDO, CSIR, Indian Statistical Institute, Calcutta, and others. Dr Dutta is now a Professor in the Department of Computer and System Sciences of the Visva Bharati University, West Bengal, India. Prior to this, he served Kalyani Government Engineering College and College of Engineering in West Bengal as a full-time faculty member. Dr Dutta remained associated as Visiting/Guest Faculty of several universities and institutes, such as West Bengal University of Technology, Kalyani University, and Tripura University. He has coauthored eight books and has also five edited books to his credit. He has published about 185 papers in various journals and conference proceedings, both international and national; as well as several book chapters in edited volumes of reputed international publishing houses like Elsevier, Springer-Verlag, CRC Press, and John Wiley, to name a few.
Dr Dutta has guided three scholars who already had been awarded their PhD. Presently, he is supervising six scholars for their PhD program. Dr Dutta has served as editor of special volumes of several international journals published by publishers of international repute such as Springer. Dr Dutta, as investigator, could implement successfully projects funded by AICTE, DST of the Government of India. Prof. Dutta has served/serves in the capacity of external member of boards of studies of relevant departments of various universities encompassing West Bengal University of Technology, Kalyani University, Tripura University, Assam University, and Silchar, to name a few. He had the opportunity to serve as the expert of several interview boards conducted by West Bengal Public Service Commission, Assam University, Silchar, National Institute of Technology, Arunachal Pradesh, Sambalpur University, and so on.
Dr Dutta is a Life Fellow of the Optical Society of India (OSI); Computer Society of India (CSI); Indian Science Congress Association (ISCA); Indian Society for Technical Education (ISTE); Indian Unit of Pattern Recognition and Artificial Intelligence (IUPRAI) – the Indian affiliate of the International Association for Pattern Recognition (IAPR); and senior member of Associated Computing Machinery (ACM); IEEE Computer Society, USA; and IACSIT.
List of Contributors
Tankut Acarman
Computer Engineering Department
Galatasaray University
Istanbul
Turkey
Loganathan Agilandeeswari
School of Information
Technology & Engineering
VIT University
Vellore
Tamil Nadu
India
Samir Kumar Bandyopadhyay
Department of Computer
Science and Engineering
University of Calcutta
Salt Lake Campus
Kolkata
West Bengal
India
Piyush Bhandari
Indian Institute of Technology Patna
Bihta, Patna
Bihar, India
S. Bharathi
Department of Electronics and Communication Engineering
Dr. Mahalingam College of Engineering and Technology, Coimbatore
Tamil Nadu, India
Sangita Bhattacharjee
Department of Computer
Science and Engineering
University of Calcutta
Salt Lake Campus
Kolkata
West Bengal
India
Siddhartha Bhattacharyya
Department of Information Technology
RCC Institute of Information Technology
Kolkata
West Bengal
India
Nesrine Bouadjenek
Laboratoire d'Ingénierie des Systèmes Intelligents et Communicants (LISIC)
Faculty of Electronics and
Computer Sciences
University of Sciences and Technology Houari Boumediene (USTHB), Algiers
Algeria
Fernando Cervantes-Sanchez
Centro de Investigación
en Matemáticas (CIMAT)
A.C., Jalisco S/N
Col. Valenciana
Guanajuato
México
Anirban Chakraborty
Department of Computer Science
Barrackpore Rastraguru
Surendranath College
Barrackpore
Kolkata
West Bengal
India
Debashish Chakravarty
Indian Institute of Technology
Kharagpur
West Bengal
India
Subhamoy Chatterjee
Indian Institute of Technology Patna
Bihta, Patna
Bihar, India
Youcef Chibani
Laboratoire d'Ingénierie des Systèmes Intelligents et Communicants (LISIC)
Faculty of Electronics and
Computer Sciences
University of Sciences and Technology Houari Boumediene (USTHB), Algiers
Algeria
Aditi Roy Chowdhury
Department of Computer
Science and Technology
Women's Polytechnic
Jodhpur Park
Kolkata
West Bengal
India
Ivan Cruz-Aceves
CONACYT – Centro de Investigación
en Matemáticas (CIMAT)
A.C., Jalisco S/N
Col. Valenciana
Guanajuato
México
Sunanda Das
Department of Computer
Science & Engineering
University Institute of Technology
The University of Burdwan
Burdwan
West Bengal
India
Supratim Das
Department of Computer
Science and Engineering
Jadavpur University
Kolkata
West Bengal
India
Sourav De
Department of Computer
Science & Engineering
Cooch Behar Government
Engineering College
Cooch Behar
West Bengal
India
Paramartha Dutta
Department of Computer Science
and System Sciences
Visva-Bharati University
Santiniketan
West Bengal
India
V. Gurunathan
Department of Electronics and Communication Engineering
Dr. Mahalingam College of
Engineering and Technology
Tamil Nadu
India
Joydev Hazra
Department of Information Technology
Heritage Institute of Technology
Kolkata
West Bengal
India
Arturo Hernandez-Aguirre
Centro de Investigación
en Matemáticas (CIMAT)
A.C., Jalisco S/N
Col. Valenciana
Guanajuato
México
Deepthi P. Hudedagaddi
School of Computer Science and Engineering (SCOPE)
VIT University
Vellore
Tamil Nadu
India
Earnest Paul Ijjina
Visual Learning and
Intelligence Group (VIGIL)
Department of Computer
Science and Engineering
Indian Institute of Technology
Hyderabad (IITH)
Hyderabad
Telangana
India
Harleen Kaur
Electrical and Instrumentation Engineering Department
Thapar University
Patiala
Punjab
India
Mahesh Kumar Kolekar
Indian Institute of Technology Patna
Bihta, Patna
Bihar, India
B. Kondalarao
Department of Mechanical Engineering
Indian Institute of Technology
Kharagpur
West Bengal
India
Kriti
Electrical and Instrumentation Engineering Department
Thapar University
Patiala
Punjab
India
Brejesh Lall
Department of Electrical Engineering
Indian Institute of Technology Delhi
India
Vaibhav Lodhi
Indian Institute of Technology
Kharagpur
West Bengal
India
Indra Kanta Maitra
Department of Information Technology
