RoboCV Module 1: Introduction to Machine Vision

2,009 views

Published on

These are the slides of the RoboCV Workshop organized by roboVITics on August 11th-12th, 2012 in TT311 Smart Classroom, VIT University, Vellore.

The workshop was delivered by the following people:
1. Mayank Prasad, President of roboVITics
2. Akash Kashyap, President of TEC - The Electronics Club of VIT
3. Akshat Wahi, Asst. Project Manager of roboVITics

Published in: Technology, Business
0 Comments
4 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
2,009
On SlideShare
0
From Embeds
0
Number of Embeds
234
Actions
Shares
0
Downloads
1
Comments
0
Likes
4
Embeds 0
No embeds

No notes for slide

RoboCV Module 1: Introduction to Machine Vision

  1. 1. Machine VisionAn IntroductionPresented to you by 1© roboVITics | Mayank Prasad, 2012 8/26/2012
  2. 2. • Image & Image Processing• Image Acquisition, Sampling and Quantization• Basic Concepts • Types of Images – Vector & Raster • Colour Space, Pixels, Resolution, Depth, Channels • Neighborhood, ConnectivityOutline 2© roboVITics | Mayank Prasad, 2012 8/26/2012
  3. 3. • 2D representation of 3D real world at any instant• Extract useful information from the image about the real world – Image Processing• Two types of images • Vector Images • Raster ImagesRaster Image – Stores images in matrix formImage 3© roboVITics | Mayank Prasad, 2012 8/26/2012
  4. 4. BASIC CONCEPTSBasic Concepts related to Image Processing 4© roboVITics | Mayank Prasad, 2012 8/26/2012
  5. 5. Digital Image – A multidimensional array of numbers Aspect Ratio – Width:Height Resolution – Width×Height Pixel – Smallest Visual Element 10 10 16 28Channel – No. of samples per point 65 70 56 43 9 6 9926703756Single Plane – Grayscale/B&W Images 32 54 96 67 78 15 256013902296Three Planes – Colour Images 21 54 47 42 67 32 158587853943 92 54 65 65 39 5 Concepts 32 65 87 99 © roboVITics | Mayank Prasad, 2012 8/26/2012
  6. 6. • Pixels are tiny little dots that form the image. They are the smallest visual elements that can be seen.• When an image is stored, the image file contains the following information: • Pixel Location • Pixel Intensity• Resolution – total number of pixels in an image• Greater resolution  Greater detail  Greater processing power requiredPixels & Resolution 6© roboVITics | Mayank Prasad, 2012 8/26/2012
  7. 7. • An image that is 2048 pixels in width and 1536 pixels in height has a total of 2048×1536 = 3,145,728 pixels or 3.1 megapixels.• One could refer to it as 2048-by-1536 or a 3.1-megapixel image.A 3.1MP Image 7© roboVITics | Mayank Prasad, 2012 8/26/2012
  8. 8. • Binary (Black & White) Image • Only two colours – black (0) & white (1) 0 1• Grayscale Image • Several shades ranging in between black and white 0 1 0 255• Colour Image • Different Colour SpacesImage Representation 8© roboVITics | Mayank Prasad, 2012 8/26/2012
  9. 9. • RGB Colour Space – Red-Green-Blue• HSV Colour Space – Hue-Saturation-Value• Y’CrCb Colour SpaceColour Spaces 9© roboVITics | Mayank Prasad, 2012 8/26/2012
  10. 10. Courtesy aishack.inRGB 10© roboVITics | Mayank Prasad, 2012 8/26/2012
  11. 11. Courtesy aishack.inHSV 11© roboVITics | Mayank Prasad, 2012 8/26/2012
  12. 12. RGB HSV• Advantages • Advantages • Intuitive • Illumination independent • Easier to use • Easier image processing • Widely used • Disadvantages• Disadvantages • Not so intuitive • Image processing is tough • Difficult to understandRGB v/s HSV 12© roboVITics | Mayank Prasad, 2012 8/26/2012
  13. 13. • Y = Luminescence or intensity• Cr = RED component minus reference value• Cb = BLUE component minus reference value• Used in video processing• Frame grabbers return images from a camera in this formatY’CrCb 13© roboVITics | Mayank Prasad, 2012 8/26/2012
  14. 14. • Depth represents the number of shades of a particular colour used in the formation of an image• Applies to grayscale as well as colour images • 1-bit : 21 = 2 shades (black & white) • 8-bit : 28 = 256 shades • 24-bit : 224 = 16,777,216 shades • 64-bit : 264 = 18,446,744,073,709,551,616 shades 8-bit 16-bit 0 1 0 1 0 255 0 65535Depth 14© roboVITics | Mayank Prasad, 2012 8/26/2012
  15. 15. • Low level task • Image Acquisition (sensing) • Preprocessing (noise reduction & enhancement)• Medium level task • Segmentation (separating regions) • Description (characteristic features) • Recognition (identify regions)• High level task • Interpretation (assign meanings)Image Processing 15© roboVITics | Mayank Prasad, 2012 8/26/2012
  16. 16. • Use Webcams, Video Cameras, Digital Cameras• Traditionally, Vidicon Camera was used• Nowadays, CCDs – Charge-Coupled Devices and CMOS Cameras are usedImage Acquisition 16© roboVITics | Mayank Prasad, 2012 8/26/2012
  17. 17. 17© roboVITics | Mayank Prasad, 2012 8/26/2012
  18. 18. Image Acquisition System 18© roboVITics | Mayank Prasad, 2012 8/26/2012
  19. 19. Sampling & Quantization 19© roboVITics | Mayank Prasad, 2012 8/26/2012
  20. 20. • D 4 D 8 8 8 4 p 4 8 p 8Neighborhood D 4 D 8 820 8© roboVITics | Mayank Prasad, 2012 8/26/2012
  21. 21. • V = (65,66,67,68,69) q 62 69 69 p 64 67 68Connectivity 65 70 7221© roboVITics | Mayank Prasad, 2012 8/26/2012
  22. 22. q n p m p and q are 8-connected m and n are m-connectedConnectivity 22© roboVITics | Mayank Prasad, 2012 8/26/2012
  23. 23. References Image Courtesy• Lectures on Robotics by • Digital Image Processing Prof. B. Seth, Mech. by Gonzalez and Woods, Engg, IIT-B (by C-DEEP) Prentice Hall• Digital Image Processing • Learning OpenCV by Gary Bradski and Adrian by Gonzalez and Woods, Kaehler, O’Reilly Media, Prentice Hall Inc.• AI Shack – • AI Shack – www.aishack.in www.aishack.inAcknowledgements 23© roboVITics | Mayank Prasad, 2012 8/26/2012
  24. 24. UP NEXT: MODULE 2Introduction to OpenCV and MATLAB 24© roboVITics | Mayank Prasad, 2012 8/26/2012
  25. 25. • Mayank Prasad President, roboVITics mayank@robovitics.in• Akshat Wahi Asst. Project Manager, roboVITics +91 909 250 3053 akshat@core.robovitics.in• Akash Kashyap President, TEC – The Electronics Club of VIT akash130791@gmail.comContacts 25© roboVITics | Mayank Prasad, 2012 8/26/2012

×