Robust range image registration: using genetic algorithms by Luciano Silva, Olga R P Bellon, Kim L Boyer

Posted by

By Luciano Silva, Olga R P Bellon, Kim L Boyer

This publication addresses the variety picture registration challenge for computerized 3D version building. the point of interest is on acquiring hugely distinct alignments among diversified view pairs of a similar item to prevent 3D version distortions; not like so much past paintings, the view pairs might show rather little overlap and needn't be prealigned. To this finish, a unique powerful evaluate metric for registration, the outside Interpenetration degree (SIM) is outlined. This degree quantifies the interleaving of 2 surfaces as their alignment is sophisticated, placing the qualitative assessment of "splotchiness," frequently utilized in connection with renderings of the aligned surfaces, onto an effective mathematical footing. The SIM is proven to be more suitable to intend squared blunders (i.e. extra delicate to positive scale adjustments) in controlling the ultimate phases of the alignment technique. The authors cross directly to mix the SIM with Genetic Algorithms (GAs) to boost a strong method for diversity photograph registration. the consequences ascertain that this method achieves exact floor registration without having for prealignment, instead of equipment in keeping with the Iterative Closest aspect (ICP) set of rules, the most well-liked thus far. Thorough experimental effects together with an intensive comparative research are offered and improved GA-based methods to enhance the registration nonetheless additional are proposed. The authors additionally improve a world multiview registration process utilizing the GA-based procedure. the implications express massive promise by way of accuracy for 3D modeling.

Show description

Read Online or Download Robust range image registration: using genetic algorithms and the surface interpenetration measure PDF

Similar algorithms and data structures books

Vorlesungen über Informatik: Band 1: Grundlagen und funktionales Programmieren

Goos G. , Zimmermann W. Vorlesungen ueber Informatik, Band 1. . Grundlagen un funktionales Programmieren (ISBN 3540244050)(de)(Springer, 2005)

Algorithms and Protocols for Wireless Sensor Networks

A one-stop source for using algorithms and protocols in instant sensor networks From a longtime overseas researcher within the box, this edited quantity offers readers with accomplished assurance of the elemental algorithms and protocols for instant sensor networks. It identifies the study that should be carried out on a few degrees to layout and investigate the deployment of instant sensor networks, and offers an in-depth research of the improvement of the subsequent iteration of heterogeneous instant sensor networks.

Algorithmic Foundations of Geographic Information Systems

This instructional survey brings jointly strains of study and improvement whose interplay provides to have major sensible influence at the region of spatial details processing within the close to destiny: geographic info platforms (GIS) and geometric computation or, extra quite, geometric algorithms and spatial info buildings.

Practical Industrial Data Networks: Design, Installation and Troubleshooting (IDC Technology (Paperback))

There are lots of information communications titles overlaying layout, deploy, and so on, yet nearly none that in particular concentrate on business networks, that are a necessary a part of the daily paintings of commercial keep an eye on platforms engineers, and the focus of an more and more huge team of community experts.

Extra resources for Robust range image registration: using genetic algorithms and the surface interpenetration measure

Example text

Also, we can apply constraint cs1 to discard incorrect correspondences, which exist between partially overlapping views. 2. 18. In this example, the alignment presents a small interpenetration stripe in the crossing region between the views. However, Surface Interpenetration Measure 39 if we analyze the angle between the corresponding points in the SIM we can determine if the surface is correctly aligned. 18. 74% Fig. 2 for (a) and (b), respectively. 2% Fig. 22 Comparison of binary images from the SIM for an incorrect alignment: (a) the visualization of the incorrect alignment; (b)-(d) results of the SIM calculated for (a) with pm1 = 5 and pm5 set to 25, 15 and 10, respectively.

The classifications are based on visual analysis totaling: 166 “aligned” views and 330 “misaligned” views. The units of MSE are in terms of range values. 6 demonstrates the results of the MSE and SIM measures for different alignments. In this experiment we computed 496 alignments obtained by exhaustive ICP-based registrations using point-topoint and point-to-plane correspondence methods [Besl and McKay (1992), Chen and Medioni (1992)]. Different combinations of adjacent views of four objects were used, totaling 124 alignments for each object.

If we analyze the number of interpenetrating points and their corresponding point distances within a very low distance range, we see that our approach has a higher fraction of interpenetrating points compared to the ICP results. 10(a) with the same analysis. 10(a) because in some regions of the alignment the surfaces are parallel. 8(d)). From the alignment obtained by ICP we sorted the points in A according to their corresponding point distance values in B and extracted the set U of non-interpenetrating points in A by the SIM calculation (the same was done for the alignment obtained by our approach).

Download PDF sample

Rated 4.30 of 5 – based on 44 votes