Título: Camera Image Mosaicing Based on an Optimized SURF Algorithm
Autores: Geng, Nan; Northwest A&F University
He, Dongjian; Northwest A&F University
Song, Yanshuang; Northwest A&F University
Fecha: 2012-12-01
Publicador: TELKOMNIKA: Indonesian journal of electrical engineering
Fuente:
Tipo: info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
Tema: No aplica
Descripción: For real-time and robust web camera image mosaicing, a method based on an optimized SURF (Speeded Up Robust Features) was proposed in this paper. Firstly, the feature points from the overlapping parts between the reference image and the target image are extracted by employing the rapid matching algorithm - BBF (Best-Bin-Fist). Then RANSAC (Random Sample Consensus) is used for mismatched features eliminating and projection matrix calculating to resample the target image and achieve the calibrated one. Finally, registered images are fused with evolutional fusion algorithm to produce image mosaicing. The results demonstrates that our image mosaicing method can stitch the images captured with the web cameras with noise and different lighting effectively and efficiently while also being robust.
Idioma: Inglés