This paper presents a image fusion technique based on PCA and fuzzy logic. the framework of the proposed image fusion technique is divided in the following major phases:
preprocesing phase
Feature extraction based on the principal component analysis The image fusion based on fuzzy set
Reconstruction final image
The figure (1) shows the framework of the proposed image fusion and its phases.
Fig. 1. The proposed approach of image fusion phases
A. Preprocessing Phase
This phase consists of three steps registration , resampling and histogram matching .In the following
1) Registration:Image fusion is the approach of combining two or more images of same scene to obtain the more informative image. The image data is recorded by sensors
on
…show more content…
Bilinear resampling is known also bilinear filtering or bilinear interpolation. Bilinear resampling is used to smooth out when they are displayed smaller or larger than they actually are. the bilinear resampling is done by interpolating between the four pixels nearest to the point that best represents that pixel
(usually in the middle or upper left of the pixel). The bilinear resampling takes a weighted average of 4 pixels in the original image nearest to the new pixel location The averaging process modifies the original pixel values and creates entirely new digital values in the output image. Bilinear resampling results are smoother,accurate,without stairstepped effect. But it has some limitations that is edges are smoothed and some extremes of the data file values are lost. It is expressed mathematically as follows Assuming i and j are integer parts of x and y, respectively; bilinear resampling is defined by:
F(x;y) =Wi;j[F(i;j)] +Wi+ 1;j[F(i+ 1;j)]
+Wi;j+ 1[F(i;j + 1)] +Wi+ 1;j+ 1[F(i+ 1;j+ 1)]
(8)
where
Wi;j= (i+ 1 x)(j+ 1 y)
Wi+ 1;j= (x i)(j+ 1 y)
Wi;j+ 1 = (i+ 1 x)(y j)
Wi+ 1;j+ 1 = (x i)(y j)
3) Histogram Matching: As previously mentioned Image fusion is the approach of combining two or more images of same scene to obtain the more informative image. The histogram matching is important step in the preprocessing for the image fusion.The histogram of an image illustrates the frequency of
These 16 features included 12 features calculated based on the 6 multispectral bands, which is mean value and standard deviation of these bands. In addition, we chose intensity, texture-variance, texture-mean, and NDVI (Normalized Difference Vegetation Index) for classifications. Finally, training samples were selected for each classification category based on the previously segmented and merged objects
In accession to the binary images, the proposed method may be tested on discrete color images also. These type of
It can then be realized that the convention of the recognition scene is employed
these algorithms to eliminate the small coefficient associated to the noise. In many signals, mostly concentration of energy is in a small number of dimensions and the coefficients of these dimensions are relatively large compared to other dimensions (noise) that has its energy spread over a large number of coefficients. In wavelet thresholding
After the image has been processed using the 3 frameworks proposed the image needs to be classified which is done using the image classification technique of image mining. Classification can be carried out by applying the method of supervised classification and unsupervised classification but here we will mainly focus upon supervised classification.
Computer Topography images are often corrupted by salt and pepper noise during image acquisition and /or transmission, reconstruction due to a number of non-idealities encountered in image sensors and communication channels. Noise is considered to be the number one limiting factor of CT image quality. A novel decision-based filter, called the wavelet based multiple thresholds switching (WMTS) filter, is used to restore images corrupted by salt-pepper impulse noise. The filter is based on a detection-estimation strategy. The salt and pepper noise detection algorithm is used before the filtering process, and therefore only the noise-corrupted pixels are replaced with the estimated central noise-free ordered mean value in the current filter window. The new impulse detector, which uses multiple thresholds with multiple neighborhood information of the signal in the filter window, is very precise, while avoiding an undue increase in
Abstract: In image processing, noise reduction and restoration of image is expected to improve the qualitative inspection of an image and the performance criteria of quantitative image analysis techniques Digital image is inclined to a variety of noise which affects the quality of image. The main purpose of de-noising the image is to restore the detail of original image as much as possible. The criteria of the noise removal problem depends on the noise type by which the image is corrupting .In the field of reducing the image noise several type of linear and non linear filtering techniques have been proposed . Different approaches for reduction of noise and image enhancement have been considered, each of which has their own limitation and advantages.
Abstract: Bad weather, such as fog, haze significantly degrade the visibility of the scene. Haze is an atmospheric phenomenon that significantly degrades the visibility of outdoor scenes. It is due to the atmosphere particles that absorb and scatter the light. Removing haze means increases the visibility of the scene and applicable for both color and gray images. The method is a fusion based strategy that derives the inputs and weight maps only from the original degraded image. To minimize the artifacts introduce the weight maps. The information of the derived inputs to preserve the regions with good visibility, by computing three measures: luminance, chromaticity, and salience
The method of feature extraction is based on the spatial distribution of the black and white pixels in the image space. We are assuming that the difference of distribution of pixels for each digit are sufficient enough to classify them.
Morphological operators take a binary image and structuring element as an input and combine them using a set operator such as union and intersection. Structuring element have dimension of 3*3 and its origin is at the center pixel. The pixels of the structuring element are compared with each pixel of an image. If it matches the condition defined by the set operator then the pixel is set to a pre-defined value (0 or 1 for binary images). There are two basic types of morphological operator, which is used in this algorithm.
Introduction:Image processing is a field that deals with manipulation of image with intent to carry out to enhance image and to extract some useful information from it. It usually deals with treating images at 2D signals and applying signal processing methods to them. It can be generally defined as a 3 step process starting by importing the image. Continuing with its analysis and ending with either an alter image or an output.
Here is a small tutorial that suffices you with the basic concepts required to put up eyes on your
Image processing refers to the construction of an image for further analysis and use. Image taken by a camera or same techniques are not actual in a form that can be used by image analysis process. The technique involves in image enhancement need to be simplified, enhanced, filtered, altered, segmented or need improvement to reducing noise, etc. Image processing is the collection of routines and techniques that alter, improve, enhance or simplify an image. Image enhancement is one of the important parts of digital image processing where image undergo for visual inspection or for machine analysis without knowledge of its source of degradation. The processes involve to bring out specific application of an image so that the result is more suitable that the original image. Image can be enhanced in various ways such as contrast enhancement, intensity, density slicing, edge enhancement, removal of noise, and saturation transformation.[1]
Recently, we developed another efficient method for matching areas in the remote sensing images using our Contourlet-based key points with the development of a simple descriptor. A matching region was formed by the convex hull of the key points matching in both images. These regions could be used for matching, fusion, and registration of remote sensing images.
Fuzzy logic is a convenient way to map an input space to an output space. Mapping input to output is the starting point for everything. Consider the following examples: