Performance Analysis of Feature based Multiple Descriptors with Histogram Equalization for Image Mosaicing

Objectives: The main objective of the proposed work is to develop an image mosaicing model for combining images different individual images. In other way, union two and evaluate performance in terms number run time seconds key features they use. Methods: this work, Histogram Equalization a processing step required make mosaic invariant intra inter intensity variability. detailed feature enhanced extracted using Scale-invariant transform (SIFT), Oriented Fast Rotated Brief (ORB), Binary Robust scalable points (BRISK) descriptors techniques. including local global are matched K-Nearest Neighbor. Then, homography performed RANdom SAmple Consensus (RANSAC) algorithm compute camera motion. Finally, warping smoothing filter size 40x40 obtain panorama image. Findings: tested on various datasets three extractors popularly used or stitching algorithms SIFT, ORB BRISK descriptors. It observed that best extractor among state-of-the-art extractors. with HE uses minimum 500 match generates panoramic shift rotation 0.0836 seconds. Novelty: models developed by researchers suffer good matching input generate mosaicing. This issue addressed multiple will provide enough matches record runtime Keywords: Mosaicing; Smoothing Filter; Equalization; Homography; Panorama Images