A New Approach to the Generation of Orthoimages of Cultural Heritage Objects—Integrating TLS and Image Data

Remote Sensing

Jakub Stefan Markiewicz¹  Piotr Podlasiak²  Dorota Zawieska²  Diego Gonzalez-Aguilera²  Fabio Remondino²  Henrique Lorenzo²  Norman Kerle² 

[1] Department of Photogrammetry, Remote Sensing and Spatial Information Systems, Faculty of Geodesy and Cartography, Warsaw University of Technology, sqr. Politechniki 1, Warsaw 00-661, Poland

关键词: plane detection;    RANSAC;    Hough transform;    CV algorithms;    TLS;    intensity orthoimages;    depth maps;    RGB orthoimages;    automation of image processing;   

DOI  :  10.3390/rs71215869

来源: mdpi

【 摘 要 】

This paper discusses the issue of automation of orthoimage generation based on Terrestrial Laser Scanning (TLS) data and digital images. The following two problems are discussed: automatic generation of projection planes based on TLS data, and automatic orientation of digital images in relation to TLS data. The majority of popular software applications use manual definitions of projection planes. However, the authors propose an original software tool to address the first issue, which defines important planes based on a TLS point cloud utilizing different algorithms (RANdom SAmple Consensus–RANSAC, Hough transform, “region growing”). To address the second task, the authors present a series of algorithms for automated digital image orientation in relation to a point cloud. This is important in cases where scans and images are acquired from different places and at different times. The algorithms utilize Scale Invariant Feature Transform(SIFT) operators in order to find points that correspond in reflectance intensity between coloure images (Red Green Blue—RGB) and orthoimages, based on TLS data. The paper also presents a verification method using SIFT and Speeded-Up Robust Features (SURF) operators. The research results in an original tool and applied Computer Vision(CV) algorithms that improve the process of orthoimage generation.

【 授权许可】

CC BY   
© 2015 by the authors; licensee MDPI, Basel, Switzerland.