Sale!

Computer Project 3: Image Enhancement

$30.00

Digital Image Processing
(ECE513)
Computer Project 3: Image Enhancement
The goal of this computer assignment is study the effects of various point and spatial
operators for contract enhancement, noise removal, edge extraction and sharpening. Use
the images on the course website.
1. Contrast Enhancement: In this part, we design histogram manipulation schemes to
improve the contrast and visual appearance of the “Pepsi” image. An important
design consideration is that the desired properties of different objects in the images
are not adversely affected.

Category:
5/5 - (2 votes)

Digital Image Processing
(ECE513)
Computer Project 3: Image Enhancement
The goal of this computer assignment is study the effects of various point and spatial
operators for contract enhancement, noise removal, edge extraction and sharpening. Use
the images on the course website.
1. Contrast Enhancement: In this part, we design histogram manipulation schemes to
improve the contrast and visual appearance of the “Pepsi” image. An important
design consideration is that the desired properties of different objects in the images
are not adversely affected. First, try histogram equalization with different number of
bins and compare the results and effects on the histogram. Devise a “desired
histogram” taking into account the above-mentioned requirement and apply
specification algorithm. Compare the resultant image and its histogram with those of
the histogram equalization and comment on their contrast enhancement abilities and
shortfalls. Explain how you arrived at your final desired histogram. MATLAB
functions for this part are: imread, imshow, imhist, histeq.
2. Spatial Filtering: Use function “imnoise” to generate a noisy version of the “Lena”
image by adding white Gaussian noise with SNR=5dB. This image should then be
used as the input to all the filters in this part of the computer assignment. Using
“conv2” apply various 2D low-pass filtering masks of different sizes (see Lecture 19)
and coefficients to remove the effects of the additive noise from the noisy image.
Compare the results of at least two filters in terms of SNR improvements and
comment on their frequency selectivity characteristics using their frequency
responses (function freqz2).
3. Median Filtering: Using the same “imnoise” function, add “salt and pepper” noise to
“Lena” image. Then apply median filters of sizes 3×3, 5×5 and 7×7 using the function
“medfilt2” to this noisy image. Increase the level of noise and determine
experimentally the maximum noise level that can be tolerated without producing
considerable blurring artifacts.
4. Edge Sharpening: Apply two different edge extraction masks covered in Lecture 20,
using function “edge” to sharpen the edges and improve the visual appearance of the
“Boat” image. Make sure edge saturation does not occur by properly selecting the
value of λ. Compare the results of these algorithms in terms of visual appearance.
1. Provide a detailed discussion on the effectiveness of these methods for image enhancement in
a report. Please carefully read the guidelines for preparing your report.

Open chat
Need help?
Hello
Can we help you?