Scribd
Upload
2 views

DIPCHnsform

Uploaded by

kailashjadhave154
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

DIPCHnsform

Uploaded by

kailashjadhave154
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

IMAGE

transform
BOUNDARY EXTRACTION

SPATIAL FILTERING
Median Filter

Applications · Noise Reduction:Removal of salt-and-pepper noise in images.

· Preprocessing: Enhancing image quality before edge detection or segmentation.

· Medical Imaging: Reducing noise in medical scans while preserving important


features like edges.

· Remote Sensing: Smoothing satellite images with high-frequency noise.

Gaussian filter is a widely used linear filter in image processing for smoothing and
noise reduction. It applies a Gaussian function to weigh pixel intensities in the spatial
do.

pplications

Noise Reduction:Common in preprocessing steps for tasks like edge detection,


segmentation, and feature extraction.

Edge DetectioOften combined with operators like Sobel or Canny for better edge
detection results (e.g., Gaussian smoothing before applying the gradient)

Medical Imaging:Smoothens scans while preserving essential details for diagnosis.

Computer Vision:Used in object detection and tracking algorithms to remove


noise

gradient filter as a high-pass filter in the spatial domain is designed to


emphasize high-frequency components in an image, such as edges and sharp
transitions. High-pass filters remove low-frequency components (smooth
regions) and retain high-frequency details (rapid intensity changes).

Applications of Gradient Filters as High-Pass Filters

1. Edge Detection:Identifies object boundaries and transitions in images.


2. Feature Extraction:Captures high-frequency features for object recognition
or classification tasks.
3. Image Enhancement:Highlights details and sharpens images.
4. Medical Imaging:Enhances visibility of structures like bones or tissue
boundarie
Frequency domain filtering involves transforming the image into the frequency
domain, manipulating its frequency components, and then transforming it back to the
spatial domain.uses for smooting and sharping of image\

2DIDEALLOWPASS FILTER

The Butterworth Low-Pass Filter (BLPF)has a sharp cutoff, the Butterworth filter
provides a more gradual transition from pass-band to stop-band, making it less prone

to artifacts like ringing in the spatial domain.

· Pass-band:Frequencies below the cutoff frequency D0D_0D0​ pass through the


filter without any attenuation (i.e., the filter has a value of 1).

· Stop-band: Frequencies above the cutoff frequency D0D_0D0​ are progressively


attenuated as the distance D(u,v)D(u, v)D(u,v) increases. The transition from the pass-
band to the stop-band is smoother compared to an ideal low-pass filter.

Gaussian Low-Pass Filter (GLPF)

Unlike the Ideal Low-Pass Filter, which has a sharp cutoff, the Gaussian filter has a
smooth, bell-shaped curve that ensures minimal artifacts like ringing in the spatial
domain.

A larger value of σ\sigmaσ leads to a wider Gaussian function, which results in a


greater attenuation of higher frequencies and a stronger smoothing effect.

Image Sharpening in the Frequency Domain involves enhancing the high-


frequency components of an image, which correspond to edges and fine details. This
process can be achieved by increasing the high frequencies and attenuating the low
frequencies.

2DIDEAL HIGH PASS FILTER


Butterworth HIGHPass Filter

HIGH FREQUECY EMPASIS FILTER

K CONTROLS THE PROPORTION OF HIGH FREQ


IMAGE

LAPLACIN FILTER IN FREQ

HIGH BOOST FILTER = a high boost filter is used to prevent low freq components
and restore details, for sharoing image, done by adding offset to image to low lfreq
component.

UNSHARP FILTER:Used for remving noiseand impulse in the image.hybrid filter.

K>1: highboost filtering

k= 1;unsharp mask

HOMOORPHIC FILTERING

Logarithmic Transformation:log(I(x,y))=log(L(x,y))+log(R(x,y))

. Fourier Transform:F(u,v)=F{log(I(x,y))}
Apply Homomorphic Filter:Ffiltered​ (u,v)=F(u,v)⋅ H(u,v)

Inverse Fourier Transform:F−1{Ffiltered​ (u,v)}=log(Ifiltered​ (x,y))

Exponentiation:Ifiltered​ (x,y)=elog(Ifiltered​ (x,y))

Ifiltered​ (x,y)

HOLE FILLING -

THCKING AND THIN=

IMAGE SEGMENATETION =

Point Detection Point detection identifies specific points in an image that have
distinct intensity values compared to their neighbors, such as corners or isolated
points.USE LAPLACIAN FILLTER

Line Detection Line detection identifies linear patterns or structures in an image.

Edge DetectionEdge detection identifies boundaries between regions with significant


intensity changes, such as object boundaries or texture discontinuities.
Roberts Operator

Prewitt Operator

Sobel Operator

Superpixels are groups of connected pixels in an image that share similar


characteristics, such as intensity, color, or texture. Instead of working with individual
pixels, which can be computationally expensive, superpixels aggregate the
information, simplifying image processing tasks while preserving important image
details.

Thresholding is a fundamental technique in image segmentation that separates


objects from the background by converting a grayscale image into a binary image. It
identifies regions of interest based on pixel intensity values
WAVELETS

WAVELET TRANSFORMATION -
The Wavelet Transform is a powerful mathematical tool used to analyze signals in
both time (or space) and frequency domains. It is particularly useful for analyzing
non-stationary signals, where frequency content changes over time (e.g., audio signals,
images, and biomedical data)
it computationally efficient.

You might also like