Assessment of image fusion procedures using entropy, image quality, and multispectral classification

Abstract

The use of disparate data sources within a pixel level image fusion procedure has been well documented for pan-sharpening studies. The present paper explores various image fusion procedures for the fusion of multi-spectral ASTER data and a RadarSAT-1 SAR scene. The research sought to determine which fusion procedure merged the largest amount of SAR texture into the ASTER scenes, while also preserving the spectral content. An additional application based maximum likelihood classification assessment was also undertaken. Three SAR scenes were tested namely, one backscatter scene and two textural measures calculated using grey level co-occurrence matrices (GLCM). Each of these were fused to the ASTER data using the following established approaches; Brovey transformation, Intensity Hue and Saturation, Principal Component Substitution, Discrete wavelet transformation, and a modified discrete wavelet transformation using the IHS approach. Resulting data sets were assessed using qualitative and quantitative (entropy, universal image quality index, maximum likelihood classification) approaches. Results from the study indicated that while all post fusion data sets contained more information (entropy analysis), only the frequency-based fusion approaches managed to preserve the spectral quality of the original imagery. Furthermore results also indicated that the textural (mean, contrast) SAR scenes did not add any significant amount of information to the post-fusion imagery. Classification accuracy was not improved when comparing ASTER optical data and pseudo optical bands generated from the fusion analysis. Accuracies range from 68.4% for the ASTER data to well below 50% for the component substitution methods. Frequency based approaches also returned lower accuracies when compared to the unfused optical data. The present study essentially replicated (pan-sharpening) studies using the high resolution SAR scene as a pseudo panchromatic band.

Fusion of multispectral and panchromatic images based on entropy and fruit fly optimization

“…In order to assess the robustness and effectiveness of the proposed method, we compared it with some existing fusion methods, i.e. PCA [18][19] the generalized IHS (GIHS) [16], the generalized IHS using spectral response functions (GIHS-SRF) [16], Wavelet Transform (WT) [6], Brovey [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20], Gram Schmidt Adaptive (GSA) [19][20][21] and FUFSER [9]. It can be seen that the spatial details extracted from the PAN image were introduced in the results obtained using all fusion methods.…”

“…Entropy has been proposed for the first time by Claude Shannon in the quantification of information. Entropy evaluates the average of the information content of an image.The fused image should be containing more information [17]. The entropy of an image is defined as following:…”

MGMDcGAN: Medical Image Fusion Using Multi-Generator Multi-Discriminator Conditional Generative Adversarial Network

“…In order to have a more accurate evaluation of the experimental results, 6 metrics are used to evaluate the fusion performance of the 10 fusion methods, including entropy (EN) [48], spatial frequency (SF), edge intensity(EI), mean gradient (MG), peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). The EN can measure the amount information contained in the fused image, SF is a metric that reflects the texture details of a image by calculating the gradient distribution, EI reflects the gradient amplitude of edge point, MG is a metric that measures the amount gradient information contained in the fused image, PSNR measures the distortion by the ratio of peak value power and noise power, and the SSIM measures the structure similarity between fused image and source images.…”

Infrared and Visible Image Fusion Based on a Latent Low-Rank Representation Nested With Multiscale Geometric Transform

“…For feasibility assessment, 32 groups of infrared and visible images of the Camp sequence were selected for experiments. By observing the image effect after fusion and calculating the entropy (EN) [25], average gradient (AvG), spatial frequency (SF) [26] and standard deviation (Std) of the source image of the sequence and the fused image, four quality evaluation indexes of the unreferenced image were comprehensively evaluated. EN is used to measure the information contained in the image, AvG is used to measure the gradient of the image, SF is used to measure the rate of change of the image gray level, and Std is used to measure the dispersion degree of the gray level of pixels.…”