Color Image Processing

Slide 1

Color Image ProcessingPresented byAnil Kumar H A[13MVD1002]1

1

Color Image ProcessingColorsimplifies object extraction and identificationhuman vision : thousands of colors vs max-24 gray levels

Color Spectrumwhite light with a prism (1966, Newton)2

Color Image Processing

RGB : Color Monitor, Color Camera, Color ScannerCMY : Color Printer, Color CopierYIQ : Color TV ,Y(luminance), I(Inphase), Q(quadrature)HSI, HSV3

Color Image ProcessingRGB Model

4

Color Image ProcessingCMY ModelColor Printer, Color CopierRGB data CMY

5

Color Image Processing

6

Color Image Processing

7

Color Image ProcessingRGB to HSI Conversion

8

Color Image ProcessingHSI to RGB Conversion

9

Image Retrieval ApplicationContent-Based Image Retrieval System

10

Image Retrieval ApplicationColor Features for Image IndexingColor Histograman estimate of the probability of occurrence of color intensitiessimple and geometric invariance(translation, rotation, and scaling)lack of spatial information of objectsDominant Colors11

Image Retrieval ApplicationExample of Color Histogram in HSI Model

Hue : range [0, 360]- Saturation : range[0, 1] - Intensity : range[0, 1] Total 36 bin quantization Hue : 6bin Saturation : 2bin Intensity: 3bin12

Introduction to Digital Image(color) ProcessingHuman vision - perceive and understand worldComputer vision, Image Understanding / Interpretation, Image processing.3D world -> sensors (TV cameras) -> 2D imagesDimension reduction -> loss of informationlow level image processing transform of one image to anotherhigh level image understandingknowledge based - imitate human cognitionmake decisions according to information in image13

13

Introduction to Digital Image Processing

Acquisition, preprocessingno intelligenceExtraction, edge joiningRecognition, interpretationintelligent14

14

Low level digital image processingLow level computer vision ~ digital image processingImage Acquisitionimage captured by a sensor (TV camera) and digitizedPreprocessingsuppresses noise (image pre-processing) enhances some object features - relevant to understanding the image edge extraction, smoothing, thresholding etc. Image segmentationseparate objects from the image backgroundcolour segmentation, region growing, edge linking etc Object description and classification after segmentation15

15

Signals and FunctionsWhat is an imageSignal = function (variable with physical meaning)one-dimensional (e.g. dependent on time)two-dimensional (e.g. images dependent on two co-ordinates in a plane)three-dimensional (e.g. describing an object in space)higher-dimensionalScalar functionssufficient to describe a monochromatic image - intensity imagesVector functions represent color images - three component colors16

16

Image FunctionsImage - continuous function of a number of variablesCo-ordinates x, y in a spatial planefor image sequences - variable (time) t Image function value = brightness at image pointsother physical quantitiestemperature, pressure distribution, distance from the observer Image on the human eye retina / TV camera sensor - intrinsically 2D2D image using brightness points = intensity imageMapping 3D real world -> 2D image2D intensity image = perspective projection of the 3D sceneinformation lost - transformation is not one-to-onegeometric problem - information recoveryunderstanding brightness info17

17

Image Acquisition & ManipulationAnalogue cameraframe grabbervideo capture cardDigital camera / video recorderCapture rate 30 frames / secondHVS persistence of vision Computer, digitised image, software (usually c)f(x,y) #define M 128 #define N 128unsigned char f[N][M] 2D array of size N*MEach element contains an intensity value

18

Image definitionImage definition:A 2D function obtained by sensing a sceneF(x,y), F(x1,x2), F(x)F- intensity, grey levelx,y - spatial co-ordinatesNo. of grey levels, L = 2BB = no. of bits

19

Brightness and 2D imagesBrightness dependent several factorsobject surface reflectance properties surface material, microstructure and markingillumination propertiesobject surface orientation with respect to a viewer and light sourceSome Scientific / technical disciplines work with 2D images directlyimage of flat specimen viewed by a microscope with transparent illuminationcharacter drawn on a sheet of paperimage of a fingerprint

20

20

Monochromatic imagesImage processing - static images - time t is constantMonochromatic static image - continuous image function f(x,y) arguments - two co-ordinates (x,y)Digital image functions - represented by matricesco-ordinates = integer numbersCartesian (horizontal x axis, vertical y axis)OR (row, column) matricesMonochromatic image function rangelowest value - black highest value - whiteLimited brightness values = gray levels21

21

Chromatic imagesColourRepresented by vector not scalarRed, Green, Blue (RGB)Hue, Saturation, Value (HSV)luminance, chrominance (Yuv , Luv)

RedGreen

Hue degrees:Red, 0 degGreen 120 degBlue 240 deg

Green

V=0S=0

22

22

Use of colour space

23

23

Image qualityQuality of digital image proportional to:spatial resolution proximity of image samples in image planespectral resolution bandwidth of light frequencies captured by sensorradiometric resolution number of distinguishable gray levelstime resolution interval between time samples at which images captured

24

24

Image summaryF(xi,yj)i = 0 --> N-1j = 0 --> M-1N*M = spatial resolution, size of imageL = intensity levels, grey levels B = no. of bits

25

Digital Image StorageStored in two partsheaderwidth, height cookie.Cookie is an indicator of what type of image filedatauncompressed, compressed, ascii, binary.File typesJPEG, BMP, PPM.26

26

PPM, Portable Pixel MapCookiePxWhere x is:1 - (ascii) binary image (black & white, 0 & 1)2 - (ascii) grey-scale image (monochromic)3 - (ascii) colour (RGB)4 - (binary) binary image5 - (binary) grey-scale image (monochromatic)6 - (binary) colour (RGB)27

27To help with understanding of systems and software design with images.

PPM examplePPM colour file RGB

P3# feep.ppm4 415 0 0 0 0 0 0 0 0 0 15 0 15 0 0 0 0 15 7 0 0 0 0 0 0 0 0 0 0 0 0 0 15 7 0 0 015 0 15 0 0 0 0 0 0 0 0 028

Image presentation (1)1.1 Image capture, representation, and storage: digital image, DPI, pixel... Example: Various quantizing level: (a) 6 bits; (b) 4 bits; (c) 2 bits; (d) 1 bit.

Image presentation (2)1.2 Color representation: Color systems: RGB, CMY/CMYK, HSI, YCbCr

SummaryWe know following terms:digital image (pixel, gray level)colormapdigitization continuous-tone imagesamplingquantizationdynamic rangespatial resolutionpixelation brightness resolutionposterization & brightness contouringdigital image processingdigital image analysis31

31

BLDescription

12Binary Image (black and white)

65464 levels, limit of human visual system

8256Typical grey level resolution