Implementation of Algorithms for the Restoration of Retinal Images

Implementation of Algorithms for the Restoration of Retinal Images

Universidad de La Salle Ciencia Unisalle Ingeniería en Automatización Facultad de Ingeniería 2016 Implementation of algorithms for the restoration of retinal images David Samir Toro Hoyos Universidad de La Salle, Bogotá Follow this and additional works at: https://ciencia.lasalle.edu.co/ing_automatizacion Part of the Robotics Commons Citación recomendada Toro Hoyos, D. S. (2016). Implementation of algorithms for the restoration of retinal images. Retrieved from https://ciencia.lasalle.edu.co/ing_automatizacion/11 This Trabajo de grado - Pregrado is brought to you for free and open access by the Facultad de Ingeniería at Ciencia Unisalle. It has been accepted for inclusion in Ingeniería en Automatización by an authorized administrator of Ciencia Unisalle. For more information, please contact [email protected]. Implementation of Algorithms for the Restoration of Retinal Images David Samir Toro Hoyos La Salle University Faculty of Engineering Automation Engineering Bogota´ 2016 Implementation of Algorithms for the Restoration of Retinal Images David Samir Toro Hoyos Degree work to obtain the title of engineer in automation Supervised by Jose´ Antonio Tumialan Borja Automation engineering La Salle University Faculty of Engineering Automation Engineering Bogota´ 2016 2 Thesis approval Jury’s signature Jury’s signature 3 Acknowledgements It is always difficult to thank enough to the people that helped me in this daunting pro- cess. Of course, this piece of work would not be possible without the blessing of God so it is for his glory. Also, My work would not be realised without the worry and uncondi- tional help of my family, my dear mother and father, Nayibe and Alonso, and brothers Ig- nacio, Jorge and Yuri. There were even kind people who offered their help when I needed it and I’ d like to thank them for accom- panying and encouraging me when my de- termination wavered and needed strengths to continue my work, to them many many thanks, it certainly was not in vain. Contents 1 Introduction 17 1.1 Case study . 17 1.2 problem statement . 17 1.3 Case objectives . 18 2 Glossary 21 2.1 Data persistence . 21 2.2 Serialization . 21 2.3 De-serialization . 21 2.4 Level . 21 2.5 Channel . 21 2.6 Histogram . 22 2.7 Intensity . 22 2.8 BGR image . 22 2.9 BGRA image . 22 2.10 Threshold . 22 2.11 Segmentation . 22 2.12 Registration . 22 2.13 Numpy Arrays . 23 2.13.1 ndarray.ndim . 23 2.13.2 ndarray.size . 23 2.13.3 ndarray.shape . 23 2.13.4 ndarray.dtype . 23 3 Art State 24 4 Methodology 29 4.1 Images in OpenCV . 29 4.2 Restoration Model . 31 4.3 Optimization Methods . 32 4.3.1 Normalization . 32 4.3.2 Process big data . 33 4 CONTENTS 5 4.3.3 Lazy evaluation . 33 4.3.4 Preventing circular references . 33 4.3.5 Multitasking . 34 4.3.6 Memory mapped files . 34 4.3.7 Memoization and caching . 34 4.4 Gathering information . 35 4.4.1 Taking samples . 36 4.5 Tests and prototyping . 36 4.6 Selecting algorithms and final application . 42 5 Algorithms design 43 5.1 General algorithms . 43 5.1.1 The area under a polygon (poligonArea) . 44 5.1.2 Overlay . 46 5.1.3 Real image – Rendered image . 47 5.1.4 Domain transformations: scaling to original results . 48 5.1.5 Normalization . 50 5.2 Load functions . 51 5.2.1 Load from sockets . 51 5.2.2 Load from URLs . 51 5.2.3 Load from files . 51 5.2.4 Load from memmaped files . 52 5.3 Pre-processing, Filters and Enhancing Methods . 52 5.3.1 Histogram equalization . 52 5.3.2 Matrix decomposition . 55 5.3.3 Smoothing with 1D-filters . 56 5.3.4 Gaussian Filter . 57 5.3.5 Bilateral Filter . 58 5.3.6 SigmoidImageFilter . 60 5.3.7 Normalized SigmoidImageFilter . 61 5.3.8 Custom filters using normS igmoid . 63 5.3.9 Sigmoid filtering and saturation . 65 5.4 Segmentations . 66 6 CONTENTS 5.4.1 Convex hull with line cuts . 67 5.4.2 Binary masks . 71 5.4.3 Alpha masks . 73 5.5 Object Recognition and Matching algorithms . 76 5.5.1 Entropy . 76 5.5.2 Histogram comparison . 77 5.5.3 Histogram matching . 77 5.5.4 Feature detection and matching . 78 5.5.5 ASIFT . 80 5.6 Using rates and probabilities . 80 5.6.1 Convexity ratio . 80 5.6.2 Rectangularity . 82 6 Implementation 88 6.1 Expert System . 101 7 User Interface 103 7.1 Command-line interface . 103 7.2 Deployment: Executable . 104 7.3 Usage . 105 8 Results 110 8.1 Result for set 1 . 111 8.2 Result for set 2 . 113 8.3 Result for set 3 . 116 8.4 Result for set 4 . 117 8.5 Result for set 5 . 118 8.6 Result for set 6 . 120 8.7 Result for set 7 . 121 8.8 Result for set 8 . 122 8.9 Result for set 9 . 124 8.10 Result for set 10 . 126 8.11 Result for set 11 . 127 8.12 Result for set 12 . 129 CONTENTS 7 8.13 Result for set 13 . 130 8.14 Result for set 14 . 131 8.15 Result for set 15 . 132 8.16 Result for set 16 . 135 8.17 Result for set 17 . 137 8.18 Result for set 18 . 138 8.19 Result for set 19 . 140 8.20 Result for set 20 . 142 8.21 Result for set 21 . 144 8.22 Result for set 22 . 146 8.23 Result for set 23 . 147 8.24 Result for set 24 . 149 8.25 Result for set 25 . 150 8.26 Result for set 26 . ..

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    288 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us