Scribd
Upload
6K views

Mini Project Abstract Model

This document discusses using a neural network to identify blur in images for restoration purposes. It begins with an abstract that notes blur identification is important for restoration and usually involves estimating the point spread function. A neural network is proposed to identify blur type and parameters from trained samples. The network uses backpropagation learning. Inverse filtering can accurately restore images if the blur is known, but adds noise, so wiener filtering provides a better tradeoff between deblurring and denoising. The project will be developed in MATLAB using image and neural network toolboxes.

Uploaded by

ggc1982
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
6K views

Mini Project Abstract Model

This document discusses using a neural network to identify blur in images for restoration purposes. It begins with an abstract that notes blur identification is important for restoration and usually involves estimating the point spread function. A neural network is proposed to identify blur type and parameters from trained samples. The network uses backpropagation learning. Inverse filtering can accurately restore images if the blur is known, but adds noise, so wiener filtering provides a better tradeoff between deblurring and denoising. The project will be developed in MATLAB using image and neural network toolboxes.

Uploaded by

ggc1982
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

TITLE <Font style: Times New Roman> <Font Size: 16> <Alignment :center>

ABSTRACT: Body <Font Style: Times New Roman> <Font size: 12> <Alignment: Justify>

ROLL NO Batch No Internal Guide ROLL NO ROLL NO Name

STUDENT NAME STUDENT NAME STUDENT NAME signature

BLUR IDENTIFICATION USING NEURAL NETWORKS FOR IMAGE RESTORATION


ABSTRACT: A prior knowledge about the distorting operator and its parameters is of crucial importance in blurred image restoration. So, the estimation of the Point Spread function is often a necessary first step in the restoration of degraded images. Usually blur refers to the low-pass distortions introduced into an image. Any type of blur, which is spatially invariant, can be expressed by the convolution in the integral equation. When the blurring operator is unknown, the image restoration becomes a blind de convolution problem. Most of the methods to solve it are iterative. So Neural Network approach is used to identification of blur. The feed forward neural network is exploited for identification of a type of blur among trained blurs and of its parameters. This network has a number of specific properties and advantages. Its back propagation learning algorithm is used for this network. This network can be used as an efficient estimator of PSF, whose precise identification is of crucial importance for the image deblurring. Inverse filtering is a very easy and accurate way to restore an image provided that we know what the blurring filter is and that we have no noise. Because an inverse filter is a high pass filter, it will tend to amplify noise as was presented in our results. But this method does not deal well with noise. To overcome this problem, we must use wiener filtering, which would trade off inverse filtering with de-noising. Wiener filtering is the optimal tradeoff between the inverse filtering and noise smoothing. This project will be developing in MATLAB 7.6, using Image Processing Tool box and Neural Network Tool box.

ROLL NO B-1 Internal Guide ROLL NO ROLL NO Mr.XXXXXXX

STUDENT NAME STUDENT NAME STUDENT NAME

You might also like