Our next presenter is Yousif Mousa Did I say that correct? Good Yousif i believe has just or very nearly submitted

his Phd in the area of Spatial Sciences and he’s been working on a project “The Canning

River in 1841 and Today” Hi everyone, i’m Yousif Mousa I’m going to talk about my project which is

“The Canning River in 1841 and Today” Thanks for my supervisor and all Curtin HIVE

and everyone who support this project The outline of this presentation will be covering

the background information of the existing survey data then i’ll introduce the survey method which

has been done 20 years ago the objective then presenting the result,

evaluation and analysis the visualisation will be and then concluding

at the end so as a background for the survey work which has been done 20 years ago so it had been done by John August Preiss for mapping the record of the Governor Hutt

of the Swan River Colony The tools that were used was Compass for measuring

the direction or bearing as well as the survey chain for measuring

distances so the available survey data includes two

field books and 14 historical maps all supported by the State Records Office so the survey method was to establish a control

point for example this point here L and M so these are here in the historical map which

is this L and M so first of all just to take the bearing from

the L to M which is here is 86 at the same time measuring the distance between

this L and M which is here 444 chain not in meters so from these after they established L and

M they measure offset from here to the change location in the river here so here this distance or offset here correspond

to the one hundred and ninety twenty nine change so do this same procedure and they also discovered

some interest point for example this is house here and trees and they do the whole Canning river so the objective will be first of all analysing

the field notes which is include survey data and historical

map then convert the field notes into a digital

map in a local system coordinate then this allow for geolocation of the the next step will be geolocation of the digital

map so it is in local system so should be geolocated to the actual system then afterwards we geolocate the historical

map based on geolocating survey data first independent accuracy assessment and further analysis and visualisation so the first step is to convert this here

this survey data here into a digital map so L the point here L is here so its given arbitrary coordinate this arbitrary coordinate its we are given it a coordinate as possible as close to the actual location

for this so with known direction from L to M and distances so we could find the location

of M and similarly we do all of the existing field into a digital map here all these have been done Magnet software so in the next step we just connect these

river boundary point from this to that to this point and go ahead to take all the

boundary points and to plot it so after we plot the river we import it in

QGIS software so its located here because we have given

it the red line because you have given it arbitrary coordinate

close as possible to the actual location so now we are geolocating the red line into

the actual coordinate system so we have found, we must find a common point

between these two system because in two different system so we have found 17 points which exists in

for example the point here assume to be point here in

the open street map not the red line the light line which is the open street map so these 17 points allow to perform the

affine and similarity projection transformation similarity has 4 parameters which is the rotation,

scale and the translation the affine transformation has 6 parameters

so with the translation there is scale in X and scale in Y as well as the rotation and the skew which

is related to the non orthogonality error so after we have done the similarity and affine

so this river is shifted as you see here its hard to control this shifted from the red line into the dark blue

line here so to assist the accuracy of the geolocating for these 17 points so the red line here is the point in the open

street map while the green points which is the survey

point are going to be shifted from the local system to the reference system so with using the affine transformation this

point is shifted for example this shifted to here this point is shifted to here so and this procedure will perform over the

whole dataset so the root mean squared error was calculated

using the affine and similarity 21.1 and 21.7 so both method have been implemented to find the geolocating of 1376 points the difference between these points using

the different techniques is about 7 in X and 3 in Y this step is about geolocating the historical

map so the historical map has labelled the point

so its easily we found 10 points which exists in the geolocated survey data

and labelled in the historical map so using them and using three different geolocating techniques similarity, projective and thin plate spline this was used in QGIS so the residual i got from similarity is about

four meters in X and Y projective is 2.2 in meters the error using the thin plate spline is zero

because this method enforce the actual point the point in the survey data to be exact position in the actual position independent accuracy assessment so we have

the overlaid the projection with the historical map generated using the projective technique the TPS method and similarity all these overlaid together so we have found some building which exists

two hundred years ago and it still exists in the historical map it is still there in the EagleView this building here is exactly here so measuring the distances from the corner the right corner here to the all the same

corner in all three generation georeferencing method so we have found there is 11.8, 13 and 14

meters using the similarity, projective and TPS method its here we could state that the similarity

generates more accurate results than the others so this building here you can see the image i don’t know the exact year for this its in

the website but this is the building here its the same which is in 1980 and this is the building today so more analysis for this project we have

found some interesting things which is related to the change in the river so looking at this open street map here the red the light line here corresponds to

the current position in the open street map the blue one which is the old bank of the

river you see here its almost the same down and

up while here is shifted to the West about 70

meters so overlaied the image with the DEM so and taking a profile from this point to

this point so you see here this profile this point the red arrow here belongs to this red point

which is which is the old part of the river while the green arrow which belongs to the

actual position for the river now so you could see clearly that its shifted

from East to West this is actually logical because if you see

the direction of the water is going from down to up so it supplied the river, the river tried

to go straight so that’s why given the time it goes to the

West end visualisation option so its this layers which is the current image taken from EagleView similarly with the historical map all overlaied

together in Unity so it allows for visual for more for different map visibilities and from different viewing directions also

allow for more analysis about any change in the river as a conclusion we have introduced a framework

for transformation of survey data whatever any projection or without any projection

into the current system for example the MGA as well as we have found there is about 21

accuracy error in the georeferencing of the survey data which is actually high but this depends on the accuracy of the 17

common points that exists between the two systems as an independent accuracy we have also found

about 13 meters positional error in the georeferencing historical map and this is actually because the river itself

its cant be in the same position so its moved somewhere also the project has been done in Virtual

Reality using the Unity software and thanks to Wesley for helping in this thanks, any questions?

nice video it was really entertaining