Upload
inquisitivo
View
1.215
Download
0
Embed Size (px)
Citation preview
IntroductionFundamentals
Image Enhancement Techniques
Speech and Image Processing LabIMAGE ENHANCEMENT TECHNIQUES
Jisa David & Seena SymonM.Tech. Signal Processing
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
INTRODUCTION
DIGITAL IMAGE & IMAGE PROCESSING
Image:2D function f(x,y) where x and y are spatial coordinatesand f is the intensity.Digital image :Representation of a 2D image as a finite set ofdigital values( pixels).Digital Image Processing:Processing a digital image by meansof a digital computer.
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
FUNDAMENTALS
STEPS IN IMAGE PROCESSING1 Image Aquisition2 Image Enhancement3 Image Restoration4 Morphological processing5 Segmentation6 Representation and Description7 Recognition
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
IMAGE ENHANCEMENT TECHNIQUES
IMAGE ENHANCEMENTProcessing of an image so that the result is more suitable than orignalimage for Specific Application.
Done in Two domain
1 Spatial Domian2 Frequency Domain
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL DOMAIN
Gray Level TransformationImage NegativesLog TransformationsPower-Law transformationPiecewise Linear Transformation
Contrast StretchingGray Level SlicingBit Plane Slicing
Spatial FilteringNeighborhood AveragingMedian Filtering
Histogram processing
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL DOMAIN
GRAY LEVEL TRANSFORMATION
Operated on individual pixels intensity values: s = T(r). r:original intensity, s: new intensity
Data independent pixel-based enhancement method.
Approaches1 Image negatives
2 Log transform
3 Power law transform
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
GRAY LEVEL TRANSFORMATION
IMAGE NEGATIVEFunction reverses the order from black to white so that theintensity of the output image decreases as the intensity of theinput increases.s= T(r) = L-1-rUsed mainly in medical images and to produce slides of the screen
Figure: (a)Original mammogram ,(b)Negative image
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
GRAY LEVEL TRANSFORMATION
LOG GRAY LEVEL TRANSFORM
s = T(r) = c log(1+r)expand dark value to enhance details of darkarea.Example:Fourier 2D Transform.
Figure: (a)Fourier Spectrum ,(b)Transformed image with c=1
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
GRAY LEVEL TRANSFORMATION
POWER LAW GRAY-LEVEL TRANSFORM
s = T(r) = crγ
Gamma correction: To compensate the built-in power lawcompression due to display characteristics. CRT intensity tovoltage response is a power function with γ=1.8-2.5.
Figure: (A)Original Image, (B)Response of monitor to A (C)GammaCorrected Image,(D)Response of monitor to C
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
GRAY LEVEL TRANSFORMATION
Figure: a) log transformation b)power-law transformation
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
GRAY LEVEL TRANSFORMATION
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
Allow more control on the complexity of T(r)Examples
Contrast stretching
Gray-level slicing
Bit-plane slicing
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
CONTRAST STRETCHINGTo increase the dynamic range of the gray levels in the imagebeing processed.The locations of (r1,s1) and (r2,s2) control the shape of thetransformation function.
If r1= s1 and r2= s2 the transformation is a linear function andproduces no changes.If r1=r2, s1=0 and s2=L-1, the transformation becomes athresholding function that creates a binary image.
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
CONTRAST STRETCHING
Figure: (A)Transformation function,(B) A low Contrast Image,(C)Result ofContrast Stretching,(D)Result of Thresholding
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
GRAY LEVEL SLICING1 To highlight a specific range of gray levels in an image (e.g. to
enhance certain features).2 The second approach is to brighten the desired range of gray
levels but preserve the background and gray-level tonalities in theimage
Figure: (a)Transformation function of (1),(b)Transformation function of (2)
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
Figure: a) original image b)Transformed image
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
BIT PLANE SLICINGTo highlight the contribution made to the total image appearanceby specific bits.
i.e. Assuming that each pixel is represented by 8 bits, the imageis composed of 8 1-bit planes.Plane 0 contains the least significant bit and plane 7 contains themost significant bit.
Application in image compression
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
PIECE-WISE LINEAR GRAY-LEVEL TRANSFORM
BIT PLANE SLICING
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
Use of spatial masks for image processing (spatial filters)Linear and nonlinear filtersLow-pass filters eliminate or attenuate high frequencycomponents in the frequency domain (sharp image details), andresult in image blurring.The basic approach is to sum products between the maskcoefficients and the intensities of the pixels under the mask at aspecific location in the image.
R = W1Z1 + W2Z2 + ......... + W9Z9 (1)
(for 3x3 filter)
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
Figure: Mechanism of Spatial Filtering
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
NEIGHBORHOOD AVERAGINGReduce the noiseAveraging using M neighborhood PixelsAs M increases, the variability of the pixel values at each locationdecreases
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
NEIGHBORHOOD AVERAGING
Figure: (a)image, (b)Corrupted Image,(c)-(f) result of averaging with M=8,16,64,128
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
MEDIAN FILTERINGThe gray level of each pixel is replaced by the median of the graylevels in the neighborhood of that pixel (instead of by the averageas before).
Median filters are nonlinear.
Median filtering reduces noise without blurring edges and othersharp details.
Median filtering is particularly effective when the noise patternconsists of strong, spike-like components. (Salt-and-peppernoise.)
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
MEDIAN FILTERING
Figure: (a)Image correpted by Salt & Pepper Noise, (b)Noise reductionwith 3x3 averaging mask, (c)Noise reduction with 3x3 median filtering.
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
SPATIAL FILTERING
EDGE ENHANCEMENTdigital image processing filter that improves the apparentsharpness of an image or videoApplication in TV broad cast and DVD
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
HISTOGRAM PROCESSING
Data-dependent pixel-based image enhancement method.
The histogram of a digital image with gray levels from 0 to L-1 isa discrete function h(rk)=nk, where:
rk is the kth gray level
nk is the No: of pixels in the image with that gray level
n is the total number of pixels in the image
k = 0, 1, 2, , L-1
Normalized histogram: p(rk)=nk/n(an estimate of theprobability of occurrence of gray level rk)
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
HISTROGRAM PROCESSING
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
FILTERING IN FREQUENCY DOMAIN
Compute Fourier transform of imageMultiply the result by a filter transfer function (or simply filter).Take the inverse transform to produce the enhanced image.
Figure: Basic Step for Filtering in the Frequency Domain
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
ENHANCEMENT IN FREQUENCY DOMAIN
Types of enhancement that can be done:Lowpass filtering: Reduce the high-frequency content – blurringor smoothingHighpass filtering: increase the magnitude of high-frequencycomponents relative to low-frequency components – sharpening.
Figure: (a)Frequency response of Low pass fillter and (b)High pass filter
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
BUTTERWORTH LOW PASS FILTERThis filter does not have a sharp discontinuity establishing a clearcutoff between passed and filtered frequencies
Transfer functiion
H(u, v) = 11+[D(u,v)/D0]2n
where D(u, v) = [(u−M/2)2 + (v −N/2)2]
where n -order of the filter,M-Number of rows,N -Number ofcolumns,D0-Cut off frequency
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
Figure: a) original image b)Noisy image c)LPF imageJisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
BUTTERWORTH HIGH PASS FILTEREdges and sharp transitions in grayvalues in an image contributesignificantly to high-frequency content of its Fourier transform.Image sharpening in the Frequency domain can be done byattenuating the low-frequency content of its Fourier transform.
Transfer functiion
H(u, v) = 11+[D0/D(u,v)]2n
where D(u, v) = [(u−M/2)2 + (v −N/2)2]
where n -order of the filter,M-Number of rows,N -Number ofcolumns,D0-Cut off frequency
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
Figure: a)Original image b)HPF image
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FREQUENCY DOMAIN
BUTTERWORTH BAND PASS FILTERBand-boost filtering boosts certain midrange freqencies andpartially corrects for blurring, but does not boost the very high(most noise corrupted) frequencies.Obtained by cascading low pass and High pass filters.Transfer functiion
H(u, v) = 1(1+[D(u,v)/D1]2n)(1+[D0/D(u,v)]2n)
where D(u, v) = [(u−M/2)2 + (v −N/2)2]
where n -order of the filter,M-Number of rows,N -Number ofcolumns,D0- Lower Cut off frequency ,D1-Upper Cut offfrequency
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
FALSE COLORING
A false-color image is an image that depicts a subject in colorsthat differ from human perception of the same subject.A false-color image is derived from a greyscale image by mappingeach pixel value to a color according to a table or function.False-color images are frequently used for viewing satelliteimages, such as from weather satellites.
Figure: (a)Original Image, (b)False Color ImageJisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES
IntroductionFundamentals
Image Enhancement Techniques
Spatial DomainFrequency Domain
REFERENCE
Rafael C. Gonzalez, richard E. Woods,Digital ImageProcessing,Pearson Education.Anil K. Jain, Fundamentals of Digital Image Processing,PrenticeHallRafael C. Gonzalez, richard E. Woods,Digital Image Processingusing MATLAB,Pearson Education.www.wikipedia.org
Jisa David & Seena Symon M.Tech. Signal Processing Speech and Image Processing Lab IMAGE ENHANCEMENT TECHNIQUES