FIG Peer Review Journal

Back

Improving the Graphical Cadastre Based on Genetic Algorithm Principles (3314)

Anna Shnaidman, Uri Shoshani and Yerach Doytsher (Israel)
Ms. Anna Shnaidman
MSc. Student
Technion
Mapping and Geoinformation Engineering
Civil and Environmental Eng. Faculty, Technion
Haifa
32000
Israel
 
Corresponding author Ms. Anna Shnaidman (email: ShnaidmanAnna[at]gmail.com, tel.: + 972 54 7827670)
 

[ abstract ] [ handouts ] [ handouts ]

Published on the web 2009-02-16
Received 2008-12-01 / Accepted 2009-02-16
This paper is one of selection of papers published for the FIG Working Week 2009 in Eilat, Israel and has undergone the FIG Peer Review Process.

FIG Working Week 2009
ISBN 978-87-90907-73-0 ISSN 2307-4086
http://www.fig.net/resources/proceedings/fig_proceedings/fig2009/index.htm

Abstract

An unconventional approach for obtaining a legally supportive, 2D coordinate based cadastre is addressed. The proposed method considers natural selection or biological optimizations known as Genetic Algorithm (GA), which has been widely applied in solving complex computation problems in a variety of disciplines. This paper describes the implementation of GA in the cadastral domain employing its principle to achieve unique, accurate and homogeneous coordinates, with a standard deviation complying with Survey of Israel (SOI) requirements. The existing method of land property registration in Israel is based on a definition of a land unit (parcel), identified by a unique number within a registered block, and graphically plotted based on field measurements linked to the national coordinate system. The products of the ground measurements are recorded in field books, following which the block borderlines, parcel turning points and additional details are plotted on field sheets. As a result, the current cadastral system is of an analog nature and deals with surface properties only. Due to an increasing number of urban centers, urban and land development projects, and the urgent need for a more accurate cadastre in the digital era, the transition to an analytical cadastre is both crucial and inevitable. Some of the research addressing this issue includes RTK GPS technologies to reinstate parcel boundaries, advance algorithms such as rubber-sheeting applied to cadastral maps, and most commonly, the Least Squares method for cadastral boundaries adjustment. All these methods are mainly analytical and straightforward. GA, on the other hand, offers a stochastic approach, which begins with a diverse range of possible solutions to a problem at hand and provides the optimal solution by mimicking the natural (biological) processes. Over a series of generations (iterations) the suggested algorithm quickly provides an encouraging and feasible solution, iteratively manipulating the initial population (randomly generated solutions). Each generation is evaluated by a fitness function, undergoing selection, mutation and recombination (crossover) to produce new and better solutions. The paper presents an implementation of GA principles in the cadastral domain. Based on a large number of simulations, and allowing for a clear resemblance between the GA solution and the conventional methods, in most cases the results of the GA are better – the coordinates are closer to their "true" value than those obtained from the common alternative, the LS technique.
 
Keywords: Geoinformation/GI; Digital cadastre; Cadastre; analytical cadastre; genetic algorithm; biological optimization

Back