Multi-Resolution Efficient Photography Image Fusion Based on Gradient Exposure

Abstract

A class of Image fusion techniques are automatically combined under different exposure level.image fusion approach is based on a gradient technique that conserves important local perceptual signals. The images that are reconstructed from integration and the gradients attain a smooth merge of the input images and at the same time possess its important features. In the proposed system the series of images are captured by digital camera which has bracketed features (Long Exposure and Short Exposure) and uses a Standard Dynamic Range (SDR) device and synthesizes an image suitable for SDR displays. The SDR device traces scene details like contrasts and gradient direction in a sequence of SDR images with dissimilar coverage levels. The depth of field is first calculated, which helps to find the distance between the nearest and farthest objects in a scene which appears sharp in an images. The scene gradient measure, luminance measure is carried out in order to measure the gradient and the contrast of image and last step is integrating the results to get the fusion result. The fusion algorithm techniques are used for fusion of images based on contrast and gradient level. This is done in a multi-resolution of brightness increase variation in the sequence. The gradient field is then integrated within dynamic range image. Experimental results prove that the proposed scheme does not need any human interaction or limit tuning for different scenes.