Non Subsampled Shearlet Transform Based Fusion of Multiple Exposure Images
- PDF / 1,078,914 Bytes
- 5 Pages / 595.276 x 790.866 pts Page_size
- 38 Downloads / 187 Views
ORIGINAL RESEARCH
Non Subsampled Shearlet Transform Based Fusion of Multiple Exposure Images Vivek Ramakrishnan1 · D. J. Pete1 Received: 26 July 2020 / Accepted: 22 September 2020 © Springer Nature Singapore Pte Ltd 2020
Abstract Fusion of multiple exposure images has attracted attention over the past decade and several algorithms have been developed, so as to capture the entire dynamic range of the scene in a single image. Capturing images with changes in exposure settings leads to a set of multiple exposure images with different areas of the scene highlighted in different image. Weak edges and fine textures of the image are lost during an under or over exposure. Also for objective evaluation we need to measure and both the structural and textural information in the images simultaneously. To address this issue an algorithm based on the Non-subsampled shearlet transform (NSST) for fusing multiple exposure images is proposed so as to depict clearly the dimly lit, brightly lit and well lit regions in a single fused image. In the proposed algorithm NSST decomposition is first performed on the images to obtain the multi-scale and multi-direction representations. The high frequency bands are fused by retaining the pixels with the highest value coefficients at each sub band at each level. Whereas the low frequency bands are fused by averaging operation. Proposed method leads to better results in visual quality. Keywords Exposure · Sub-sampling · Shearlet · Fusion · Multi-scale · Multi-direction · Averaging · Visual
Introduction Majority of the images use one byte per pixel for representing each of the R, G and the B components. This gives 256 values for representing each pixel. The following scene of the general configuration of an indoor scene with a window is quite common. This will lead to both bright and dark regions in the same scene as shown in Fig. 1. Images are taken with multiple exposures and combined using some fusion rule, which enables us to highlight wellexposed regions of the image in specific shots and the combined/fused images contain all the regions well-lit or properly exposed. Mertens et.al. [1] have developed an algorithm called as exposure fusion that combines multiple exposure images into a single image based on certain parameters viz. the Saturation, Contrast and Well-exposedness. This approach basically focusses on obtaining weights using the three parameters and performing a weighted sum of pixels * Vivek Ramakrishnan [email protected] 1
Department of Electronics Engineering, Datta Meghe College of Engineering, Sector‑3, Airoli, Navi Mumbai 400708, India
to derive the resultant image. The non-subsampled shearlet transform (NSST) based approach for the fusion of multiple exposure images, described in this work is an effort to fuse the images in the transform domain and perform the inverse transform to obtain the result in the spatial domain. Many of the current approaches for fusing multiple exposure images is mainly in the spatial domain, a Low-Dynamic range solution to the
Data Loading...
