Image Processing Ppt

SEMINAR REPORT SEMINAR REPORT ONON

IMAGE PROCESSING IMAGE PROCESSING TECHNIQUESTECHNIQUES

CONTENTSCONTENTS

Introduction Introduction Various techniques of image processing Various techniques of image processing Geometric TransformationGeometric Transformation Image SmoothingImage Smoothing Contrast EnhancementContrast Enhancement Negative TransformationNegative Transformation Edge DetectionEdge Detection Conclusion Conclusion

INTRODUCTIONINTRODUCTION IMAGE:-IMAGE:-

Image is a optical appearance of object produced Image is a optical appearance of object produced through mirror or lenses .through mirror or lenses .

DIGITAL IMAGE:-DIGITAL IMAGE:-

Digital image is composed of finite no. of Digital image is composed of finite no. of elements each of which has a particular location elements each of which has a particular location and intensity values.and intensity values.

DIGITAL IMAGE PROCESSING:-DIGITAL IMAGE PROCESSING:-

The field of digital image processing refers to The field of digital image processing refers to processing of digital images by means of digital processing of digital images by means of digital computers .computers .

VARIOUS TECHNIQUES OF VARIOUS TECHNIQUES OF IMAGE PROCESSINGIMAGE PROCESSING

Geometric TransformationGeometric Transformation Image SmoothingImage Smoothing Contrast EnhancementContrast Enhancement Edge DetectionEdge Detection Negative TransformationNegative Transformation Morphological OperationMorphological Operation

GEOMETRIC GEOMETRIC

TRANSFORMATIONTRANSFORMATION

GEOMETRIC GEOMETRIC TRANSFORMATIONTRANSFORMATION

Geometric transformation are simply alignments Geometric transformation are simply alignments done on images to collect the exact appearance of done on images to collect the exact appearance of an objects.an objects.

(i) Translation(i) Translation

(ii)Rotation(ii)Rotation

(iii)Scaling(iii)Scaling

TranslationTranslation

In translation, axes of individual are displayed by In translation, axes of individual are displayed by their respective displacement factor.their respective displacement factor. XT=X+X0 YT=Y+Y0 XT=X+X0 YT=Y+Y0 ZT=Z+ZOZT=Z+ZO

Input imageTranslated image

RotationRotation

Rotation is one type of alignment where an image is rotated w.r.t different axes with different angles to align with the image plane

Input image Output image

ScalingScaling

In scalling operation axes of indivisuals are multiplied In scalling operation axes of indivisuals are multiplied by respective scaling factor.by respective scaling factor.

XXSS=S=SXXX YX YSS=S=SYYY ZY ZSS=S=SZZZZ

Input imageOutput image scaledBy 0.5

IMAGE IMAGE

SMOOTHINGSMOOTHING

IMAGE SMOOTHINGIMAGE SMOOTHING

It is a technique to reduce noise in a digital It is a technique to reduce noise in a digital image. It is usually applied to diminish the image. It is usually applied to diminish the spurious noise.spurious noise.

Low pass filteringLow pass filtering Gaussian filteringGaussian filtering Median filtering Median filtering

Low pass FilteringLow pass Filtering

Since noise contribute in high frequency component, Since noise contribute in high frequency component, so LPF is used to suppress such high frequency so LPF is used to suppress such high frequency component to remove the noise.component to remove the noise.

General imageGeneral image Output image

Cont…Cont…

Image with Gaussian noiseImage with Gaussian noise Output of LPF

Cont…Cont…

Image with salt and pepper noise Output of LPF

Gaussian FilteringGaussian FilteringIt is sometimes useful to apply a Gaussian smoothing filter to an image before performing edge detection. The filter can be used to soften edges, and to filter out spurious points(noise) in an image.

General image Output of Gaussian filter

Cont…Cont…

Image with Gaussian noise Output of Gaussian filter

Cont…Cont…

Image with salt and pepper noise Output of Gaussian filter

Median FilteringMedian FilteringThis is a ordered statistic filter is defined by the equation:

wi =1 if i= (Ω-1)/2 =0 otherwise

Where Ω is an odd number Median filter is very effective to remove impulsive noise.

General image Output of median filter

Cont…Cont…

Image with Gaussian noise Output of median filter

Cont…Cont…

Image with salt and pepper noise Output of median filter

CONTRASTCONTRAST

ENHANCEMENTENHANCEMENT

CONTRAST ENHANCEMENTCONTRAST ENHANCEMENT

Processing an image means to enhance certain Processing an image means to enhance certain features of the image.features of the image.

One of the defect found in images is its poor One of the defect found in images is its poor contrast due to inadequate lighting, aperture contrast due to inadequate lighting, aperture size, shutter speed etc.size, shutter speed etc.

HistogramHistogram:-It gives global description of an :-It gives global description of an image.image.

Histogram equalization:-Histogram equalization:-It means increasing It means increasing dynamic range of the gray level.dynamic range of the gray level.

Cont…Cont…

p(rk)=nk/n

Low contrast image Histogram of low contrast image

Cont…Cont…

Output image Histogram of output image

EDGEEDGE

DETECTIONDETECTION

EDGE DETECTIONEDGE DETECTION

It s the most common approach to detect the It s the most common approach to detect the discontinuity.discontinuity.

Edges are the boundary between two regions Edges are the boundary between two regions having distinct gray levels.having distinct gray levels.

Ariel images,road secton,river etc. need Ariel images,road secton,river etc. need boundary information. boundary information.

Cont…Cont…

There are four operators for edge detection:-There are four operators for edge detection:- Prewitt operatorsPrewitt operators Robert operatorsRobert operators Canny edge detectionCanny edge detection

Prewitt OperatorPrewitt Operator

PrewittPrewitt is a method of is a method of edge detection in in image processing which calculates the which calculates the maximum response of a set of maximum response of a set of convolution kernels to find the local edge to find the local edge orientation for each orientation for each pixel..

Mask of prewitt edge detectorMask of prewitt edge detector-1 -1 -1

0 0 0

1 1 1

-1 0 1

-1 0 1

-1 0 1

Gy= Gx=

Cont…Cont…

Input image Output image

Robert OperatorRobert Operator The equivalent masks of the robert operator areThe equivalent masks of the robert operator are

d1=g0 – g1 and d2= g0 – g3d1=g0 – g1 and d2= g0 – g3 0 -1

0 1

0 0

-1 1

INPUT IMAGE OUTPUT IMAGE

Canny OperatorCanny Operator

Canny operator is used for wide range of edge Canny operator is used for wide range of edge detection.detection.

It smooth the image by Gaussian filter to reduce noise, It smooth the image by Gaussian filter to reduce noise, unwanted detail and texture.unwanted detail and texture.

Canny then compute gradient using gradient operatorsCanny then compute gradient using gradient operators

Cont…Cont…

Input image Output image

NEGATIVENEGATIVE

TRANSFORMATIONTRANSFORMATION

NEGATIVE TRANFORMATIONNEGATIVE TRANFORMATION

Negative transformation means reversing the Negative transformation means reversing the intensity levels of an image.intensity levels of an image.

This type of processing is used for enhancing This type of processing is used for enhancing white or gray detail embedded in dark region white or gray detail embedded in dark region of an image.of an image.

Mostly used in medical application where Mostly used in medical application where minute informations are important.minute informations are important.

Cont..Cont..

S=L-1-rS:-Negative of an imageL:-total no.of pixelsR:-max intensity value

Input image Negative of input image

Cont…Cont…

OUTPUT IMAGE INPUT IMAGE

CONCLUSIONCONCLUSION The seminar “Image processing technique” has been successfully finished. The seminar “Image processing technique” has been successfully finished.

we perform various operation:-we perform various operation:- Transformation of imageTransformation of image Smoothing of imageSmoothing of image Contrast enhancementContrast enhancement Negative transformationNegative transformation Edge detectionEdge detection

Now a days image processing have great Now a days image processing have great application .These techniques are used in military purpose ,for identifying application .These techniques are used in military purpose ,for identifying certain regions ,hills ,rivers. Image processing have many applications in certain regions ,hills ,rivers. Image processing have many applications in medical purposes, specially the image negative for identifying small medical purposes, specially the image negative for identifying small defects in the body.defects in the body.