Nguyễn Công Phương
DIGITAL IMAGE PROCESSING
Image Restoration and Blind Deconvolution
Contents
Introduction to Image Processing & Matlab Image Acquisition, Types, & File I/O Image Arithmetic
Image Transform Spatial & Frequency Domain Filter Design
Binary Image Processing Image Encryption & Watermarking Image Classification & Segmentation
I. II. III. IV. Affine & Logical Operations, Distortions, & Noise in Images V. VI. VII. Image Restoration & Blind Deconvolution VIII. Image Compression IX. Edge Detection X. XI. XII. XIII. Image – Based Object Tracking XIV. Face Recognition XV. Soft Computing in Image Processing
sites.google.com/site/ncpdhbkhn 2
Image Restoration and Blind Deconvolution
1. Introduction 2. Image Representation 3. Deconvolution
sites.google.com/site/ncpdhbkhn 3
Introduction
• Image enhancement (improving image quality
for visual purposes) includes: – Noise removal, – Constrast stretching, – Histogram equalization, – Filtering, – Restoration through deconvolution.
sites.google.com/site/ncpdhbkhn 4
Image Representation (1)
g m n , )
(
f H
m n ( , )
f m n , )
(
Σ
Defection function H(m, n, θ)
f H
( )k
sites.google.com/site/ncpdhbkhn 5
Image Representation (2)
Motion – blurred
Out-of-focus blurred
sites.google.com/site/ncpdhbkhn 6
Image Restoration and Blind Deconvolution
1. Introduction 2. Image Representation 3. Deconvolution
a. Lucy – Richardson Method b. Wiener Method c. Blind Method
sites.google.com/site/ncpdhbkhn 7
Deconvolution
• The reverse of convolution. • A.k.a. “inverse filtering”.
Fourier transform
g
h
f
G HF
1
f
F D G
Inverse Fourier transform
• 3 main types of deconvolution algorithms:
– Lucy – Richardson method, – Wiener method, – Blind method.
sites.google.com/site/ncpdhbkhn 8
Lucy – Richardson Method
t
1t
f
f
t
h
h
g f
sites.google.com/site/ncpdhbkhn 9
Wiener Method
W
* H F 2
H F N
sites.google.com/site/ncpdhbkhn 10
Blind Deconvolution
sites.google.com/site/ncpdhbkhn 11
