A structural patch decomposition approach for MME- Image fusion technique using video

Abstract

Removal of shadows from one image could be a difficult drawback. Manufacturing a high-quality shadow-free image that is indistinguishable from a replica of a real shadow-free scene is even tougher Shadows in pictures area unit generally full of many phenomena within the scene, as well as physical phenomena like lighting conditions, kind and behavior of shadowy surfaces, occluding objects, etc. Additionally, shadow regions might endure post acquisition, image process transformations, e.g., distinction sweetening, which can introduce noticeable artifacts within the shadow-free pictures. We dispute that the assumptions introduced in most studies arise from the quality of the matter of shadow removal from one image and limit the category of shadow pictures which might be handled by employing a Modified Multi-exposure image fusion (MMEF) technique. Experimental results showing definitively the capabilities of our algorithmic rule are given. The difference is that HDR reconstruction works in the radiance domain where the value is linear w.r.t. the exposure, while MMEF works in the intensity based domain. Compared with object motion, camera motion is relatively easy to tackle via either setting a tripod or employing some registration techniques.