Open Cv – An Introduction To The Vision


Published on

Open Cv – An Introduction To The Vision library

Published in: Education, Technology, Business
  • Be the first to comment

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Open Cv – An Introduction To The Vision

  1. 1. OPEN CV – AN INTRODUCTION TO THE VISION LIBRARY AND ITS APPLICATIONS Hemanth Haridas Co founder VidPulp Technologies
  2. 2. WHAT IS THIS TALK ABOUT? <ul><li>An basic introduction to Open CV, a cross platform vision library. </li></ul><ul><li>To get students interested in Vision and related fields. </li></ul><ul><li>Introduction to the capabilities of Open CV. </li></ul><ul><li>Applications of the library. </li></ul><ul><li>Limitations of the library </li></ul>
  3. 3. WHAT IS THIS TALK NOT ABOUT? <ul><li>Not a programmatic tutorial on Open CV. </li></ul><ul><li>Not a detailed explaination about Open CV. </li></ul><ul><li>No math behind Open CV covered. </li></ul>
  4. 5. COMPUTER VISION <ul><li>Rapidly growing field because of cheaper and more capable cameras and affordable processing power. </li></ul><ul><li>Vision algorithms are starting to mature. </li></ul><ul><li>Open CV has helped computer vision grow as a field </li></ul><ul><li>Open CV helps jump-start research by providing them with a computer vision and machine learning infrastructure </li></ul>
  5. 6. CONTENTS <ul><li>Introduction </li></ul><ul><li>Who Uses OpenCV? </li></ul><ul><li>What is Computer Vision? </li></ul><ul><li>How complex is the problem? </li></ul><ul><li>Details about the Library. </li></ul><ul><li>Summary </li></ul>
  6. 7. INTRODUCTION <ul><li>OpenCV ( Open Source C omputer V ision) </li></ul><ul><li>library of programming functions for real time computer vision. (computer efficiency) </li></ul><ul><li>Written in optimized C and C++ </li></ul><ul><li>Runs in windows, Linux and MAC OS. </li></ul><ul><li>Can develop in C, C++, python, ruby, matlab </li></ul>
  7. 8. INTRODUCTION <ul><li>Simple to use infra </li></ul><ul><li>Helps to build fairly complicated applications. </li></ul><ul><li>500 functions spanning </li></ul><ul><ul><li>Factory product inspection </li></ul></ul><ul><ul><li>medical imaging </li></ul></ul><ul><ul><li>Security </li></ul></ul><ul><ul><li>user interface </li></ul></ul><ul><ul><li>camera calibration </li></ul></ul><ul><ul><li>stereo vision </li></ul></ul><ul><ul><li>robotics </li></ul></ul><ul><li>contains a full, general-purpose Machine Learning Library (MLL) </li></ul>
  8. 9. WHO USES OPEN CV? <ul><li>Surveillance </li></ul><ul><li>images and video on the Web. (Flickr, picasa face recognition and tagging) </li></ul><ul><li>aerial and street-map images (such as in Google’s Street View) make heavy use of camera calibration and image stitching techniques </li></ul><ul><li>safety monitoring, unmanned flying vehicles, or biomedical analysis. </li></ul><ul><li>manufacturing: virtually everything that is mass-produced has been automatically inspected at some point using computer vision. </li></ul>
  9. 10. WHO USES OPEN CV? <ul><li>license for OpenCV has been structured such that you can build a commercial product using all or part of OpenCV. </li></ul><ul><li>You are under no obligation to opensource </li></ul><ul><li>large user community that includes people from major companies (IBM, Microsoft , Intel, SONY, Siemens, and Google, to name only a few) and research centers (such as Stanford, MIT, CMU, Cambridge, and INRIA). </li></ul>
  10. 11. WHO USES OPEN CV? <ul><li> - 20,000 members </li></ul><ul><li>OpenCV was a key part of the vision system in the robot from Stanford, “Stanley”, which won the $2M DARPA Grand Challenge desert robot race </li></ul><ul><li>web maps, image scan alignment, medical image noise reduction, object </li></ul><ul><li>analysis, security and intrusion detection systems, automatic monitoring and safety systems, </li></ul><ul><li>manufacturing inspection systems, camera calibration, military applications, and </li></ul><ul><li>unmanned aerial, ground, and underwater vehicles </li></ul>
  11. 12. WHAT IS COMPUTER VISION? <ul><li>transformation of data from a still or video camera into either a decision or a new representation. </li></ul><ul><li>Data - > “the camera is mounted in a car” or “laser range fi nder indicates an object is 1 meter away”. </li></ul><ul><li>Decision -> “there is a person in this scene” or “there are 14 tumor cells on this slide” </li></ul><ul><li>new representation -> turning a color image into a grayscale image </li></ul>
  12. 13. HOW HARD CAN THAT BE? <ul><li>Human brain divides the vision signals to many channels. </li></ul><ul><li>Identifies important parts </li></ul><ul><li>Complex feedback mechanism that is little understood. </li></ul><ul><li>draw on cross-associations made from years of living in the world. </li></ul><ul><li>Controls lighting through the iris. </li></ul>
  13. 14. HOW A MACHINE SEES IT? <ul><li>A 2d Image of a 3d object. </li></ul><ul><li>No definite way to reconstruct the 3d image. </li></ul>
  14. 15. HOW A MACHINE SEES IT? <ul><li>Images are corrupted by noise and distortions. </li></ul><ul><li>(weather, lighting, reflections, movements) </li></ul><ul><li>Additional contextual knowledge is used </li></ul><ul><li>Which is helpful in matching </li></ul>
  15. 16. NOISE <ul><li>Edge Detection -> impossible to detect edges by comparing a point to its neighbours </li></ul><ul><li>If the comparison is made over a localized area its easier. </li></ul><ul><li>Compensating noise by using statistics over time. </li></ul><ul><li>Explicit models learnt from available data. (lens distortions) </li></ul>
  16. 17. CONTEXTUAL INFORMATION <ul><li>The decision taken by vision algorithms depend on the application it is used for. </li></ul><ul><li>Security system that alerts if a person tries to cross a fence. </li></ul><ul><li>monitoring system that counts how many people cross through an area in an amusement park. </li></ul><ul><li>Strategy for vision algos in security cameras different from that of in robots. </li></ul><ul><li>The more constrained our context , the better the solution will be. </li></ul>
  17. 18. ABOUT THE LIBRARY <ul><li>aimed at providing the basic tools needed to solve computer vision problems </li></ul><ul><li>high-level functionalities in the library will be sufficient to solve the more complex problems in computer vision </li></ul><ul><li>the basic components in the library are complete enough to enable creation of a complete solution </li></ul><ul><li>After you develop a first draft solution, check for weakness and fix it. </li></ul>
  20. 21. RESOURCES <ul><li>Download Open CV - </li></ul><ul><li>Install guide and tutorials - </li></ul><ul><li>IDE – eclipse, .net , VC++ </li></ul>
  21. 22. SUMMARY <ul><li>Open CV is a open source library to implement Computer vision algorithms </li></ul><ul><li>Computer vision is a complex problem. </li></ul><ul><li>Made easy with enough context information. </li></ul><ul><li>Computer vision is a interesting and a fast growing field and skills in this field is niche , in demand </li></ul>
  22. 23. OCR ALGORITHMS <ul><li>Optical character recognition, to recognize text in scanned documents. </li></ul><ul><li>Useful in detecting text in videos </li></ul><ul><li>For extracting contextual information in videos. </li></ul><ul><li>Tesseract and GOCR – open source OCRs available </li></ul>
  23. 24. SPEECH TO TEXT ALGOS <ul><li>Algorithms to convert from Speech to text. </li></ul><ul><li>Text to speech conversion algos are available. </li></ul><ul><li>Language translation research is underway. </li></ul>
  24. 25. QUESTIONS?
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.