Lecture Digital image processing: Image restoration and blind deconvolution - Nguyễn Công Phương

Lecture Digital image processing - Image restoration and blind deconvolution presents the following content: Image representation, deconvolution (Lucy – Richardson method, Wiener method, blind method).
Nội dung trích xuất từ tài liệu:
Lecture Digital image processing: Image restoration and blind deconvolution - Nguyễn Công Phương Nguyễn Công Phương DIGITAL IMAGE PROCESSING Image Restoration and Blind Deconvolution Contents I. Introduction to Image Processing & Matlab II. Image Acquisition, Types, & File I/O III. Image Arithmetic IV. Affine & Logical Operations, Distortions, & Noise in Images V. Image Transform VI. Spatial & Frequency Domain Filter Design VII. Image Restoration & Blind Deconvolution VIII. Image Compression IX. Edge Detection X. Binary Image Processing XI. Image Encryption & Watermarking XII. Image Classification & Segmentation 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 ) Defection function H(m, n, θ) Σ f H  (k ) sites.google.com/site/ncpdhbkhn 5 Image Representation (2) Motion – blurred sites.google.com/site/ncpdhbkhn Out-of-focus blurred 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”. g h f  Fourier transform  G  HF F  D 1G  Inverse Fourier transform  f • 3 main types of deconvolution algorithms: – Lucy – Richardson method, – Wiener method, – Blind method. sites.google.com/site/ncpdhbkhn 8 Lucy – Richardson Method t 1  g  f  f h t t   h  f  sites.google.com/site/ncpdhbkhn 9 Wiener Method * H F W 2 H FN sites.google.com/site/ncpdhbkhn 10 Blind Deconvolution sites.google.com/site/ncpdhbkhn 11