17. Optimal Image Fusion using Neuro-Fuzzy Algorithm and SVM

International Journal of Advance Engineering and Research Development

In remote sensing community, different image sensors provide different data with different spectral and spatial resolution. Multispectral imaging sensors collect poor spatial resolution multispetral data, while panchromatic imaging ones provide adequate spatial resolution panchromatic data. The fusion of images is the process of combining two or more images into a single image retaining important features from each. Ideally, the method used to merge data sets with high-spatial and high-spectral resolution should not distort the spectral characteristics of the high-spectral resolution data. This paper compares the results of two different methods based on the curvelet and wavelet transforms. The central idea of all image fusion methods based on multiresolution analysis is to extract the panchromatic image the spatial detail that is not present in the multispectral image in order to insert it later in the latter. These fusion methods, as well as curvelet and wavelet based methods have been used to merge Alsat 2 algerian satellite images with a ratio 4 : 1. We have estimated the validity of each fusion method by analyzing, visually and quantitatively, the quality of the resulting fused images.

6th International Symposium on Telecommunications (IST), 2012

The information content of a single image is mainly limited by the spatial and spectral resolution of the imaging system. Current imaging systems offer a trade-off between high spatial and high spectral resolutions. No single system offers both of these characteristics. In order to obtain the both characteristics in a single image, that is high spatial and spectral resolutions, image fusion technique can be employed. Image fusion is a process of combining two or more images into an image. It can extract features from source images, and provide more information than one image can. In this research, we propose a novel method for multimodality remote sensing image fusion. As a possible remedy for this problem we propose a technique for the fusion of Panchromatic and Multispectral images based on Curvelet transform. This paper introduces new application of the curvelet in multispectral remote sensing image fusion. The methodological approaches proposed in this paper result in merged images with improved quality with respect to those obtained by IHS, DWT, wavelet à trous algorithm and ridgelet and curvelet transform. Results show proposed method preserves more spectral features with less spatial distortion.

This paper displays a novel picture combination strategy appropriate for container honing of multispectral (MS) groups, taking into account multi-determination examination. The low-determination MS groups are honed by infusing high-pass directional subtle elements extricated from the high-determination panchromatic (Pan) picture by method for the Curvelet change, which is a non-distinguishable MRA, whose premise capacity are directional edges with dynamically expanding determination Recently, a few studies demonstrated that wavelet-based picture combination strategy gives high caliber of the ghastly substance of the intertwined picture. Be that as it may, the greater part of wavelet-based techniques have a spatial determination of the intertwined result not exactly the Brovey, IHS, and PCA combination strategies. In this paper, another technique in view of the Curvelet change utilizing SVM is presented which speaks to preferable edges over wavelets. Since edges are vital in picture understanding, one great approach to improve spatial determination is to upgrade the edges. Curvelet-based picture combination technique gives wealthier data in the spatial and ghostly areas at the same time. Picture combination is performed utilizing Curvelet Transform with SVM Algorithm. This new strategy has achieved an ideal combination result. For the execution of this proposed work we utilize the Image Processing Toolbox under Matlab Software.

International Journal of Advanced Research in Electrical, Electronics and Instrumentation Energy, 2015

Image fusion has found many applications in computer vision, remote sensing, intelligent robots and military purposes. The use of different image fusion algorithms gives precise resultant images.In many remote sensing applications, the quantity of image data from the satellite sensors has been increasing because of advanced sensor technology.To avoid the limitations of single sensor images, multisensory image fusion provides the data that is suitable for further applications by eliminating the problem of lack of information. In this paper, a literature review has been made based on different techniques for combining multispectral images available. It includes IHS transform, High Pass filtering, PCA analysis, ANN, Wavelet transform and DCT. One of the effective techniques to get a good quality image is by using the Fuzzylet Fusion Algorithm in which the advantages of both Stationary Wavelet Transform and Fuzzy logic are combined.

Current Science

This study proposes a new multispectral (MS) and panchromatic (PAN) image fusion algorithm based on regionally weighted principal component analysis (RW-PCA) and wavelet. First, the MS images are segmented into spectrally similar regions based on the fuzzy c-means (FCM) clustering method. Secondly, based on the spectral vector's degree of membership in each region, a new RW-PCA method is proposed to fuse the MS and PAN images region by region, and fused MS images are obtained. In the traditional PCA-based fusion method, the MS and PAN images are fused globally with the same transform method. In the proposed RW-PCA-based fusion method, the local spectrum information of the MS images is employed, and the spectral information is better preserved in the fused MS images. Finally, in order to improve the quality of spectral and spatial details, the above fused MS images and the original PAN images are further fused using the wavelet-based fusion method, and the final fused MS images are obtained. Experimental results demonstrated that the proposed image fusion algorithm performs better in spectral preservation and spatial quality improvement than some other methods do.

2010 10th International Conference on Hybrid Intelligent Systems, 2010

In recent years, multispectral image fusion methods are viewed as an effective tool to analyze multiband remote sensing images. In this paper a novel hybrid multispectral image fusion method using combine framework of wavelet transform and fuzzy logic is proposed. The proposed method provides novel tradeoff solution between the spectral and spatial fidelity and preserves more detail spectral and spatial information. New hybrid image fusion rules are also proposed. Proposed method is applied on registered Panchromatic and Multispectral images and simulation results are compared with standard image fusion parameters. The simulation results of proposed method also compared with five different standard Pan sharpening methods available in literature. It has been observed from simulation results that proposed algorithm preserves better spatial and spectral information and better visual quality compared to earlier reported methods.

International Journal of Engineering Sciences & Research Technology, 2013

The fusion of high-spectral/low-spatial resolution multi panchromatic satellite images is a very useful technique in various applications of remote sensing. Recently, some studies showed that a wavelet-based image fusion method provides high quality spectral content in fused images. However, most wavelet-based methods yield fused results with spa the Brovey, IHS, and PCA fusion methods. In this paper, we introduce a new method based on a curvelet transform, which represents edges better than wavelets. Since edges play a fundamental role in image effective means to enhance spatial resolution is to enhance the edges. The curvelet provides richer information in the spatial and spectral domains simultaneously. We performed Landsat ETM+ image fusion and found that the proposed method provides optimum fusion results.

Optical …, 2007

A new methodology for fusing satellite sensor imagery, based on tailored filtering in the Fourier domain is proposed. Finite-duration Impulse Response (FIR) filters have been designed through an objective criterion, which depends on source image characteristics only. The designed filters allow a weighted fusion of the information contained in a fine spatial resolution image (PAN) and in a multispectral image (MULTI), respectively, establishing a trade-off between spatial and spectral quality of the resulting fused image. This new technique has been tested with Landsat Enhanced Thematic Mapper Plus (ETM+) imagery. Spatial and spectral quality of the fused images was compared with the results provided by Mallat's Wavelet algorithm. The images fused by the proposed method were characterized by a spatial resolution very close to the PAN image, and by the spectral resolution of the MULTI image.

Acta Universitatis Sapientiae, Informatica

Image fusion is a method of combining the Multispectral (MS) and Panchromatic (PAN) images into one image contains more information than any of the input. Image fusion aim is to decrease unknown and weaken common data in the fused output image at the same time improving necessary information. Fused images are helpful in various applications like, remote sensing, computer vision, biometrics, change detection, image analysis and image classification. Conventional fusion methods are having some side effects like assertive spatial information and uncertain color information is an usually the problem in PCA and wavelet transform based fusion is a computationally in depth process. In order to overcome these side effects and to propose alternative soft computing fusion approach for conventional fusion methods we exploit image fusion using fuzzy logic technique to fuse two source images obtained from different sensors to enhance both spectral and spatial information. The proposed work here ...

Image fusion based on the Fourier and wavelet transform methods retain rich multispectral details but less spatial details from source images. Wavelets perform well only at linear features but not at non linear discontinuities because they do not use the geometric properties of structures. Curvelet transforms overcome such difficulties in feature representation. In this paper, we define a novel fusion rule via high pass modulation using Local Magnitude Ratio (LMR) in Fast Discrete Curvelet Transforms (FDCT) domain. For experimental study of this method Indian Remote Sensing (IRS) Resourcesat-1 LISS IV satellite sensor image of spatial resolution of 5.8m is used as low resolution (LR) multispectral image and Cartosat-1 Panchromatic (Pan) of spatial resolution 2.5m is used as high resolution (HR) Pan image. This fusion rule generates HR multispectral image at 2.5m spatial resolution. This method is quantitatively compared with Wavelet, Principal component analysis (PCA), High pass filtering(HPF), Modified Intensity-Hue-Saturation (M.IHS) and Grams-Schmidth fusion methods. Proposed method spatially outperform the other methods and retains rich multispectral details.