Adjustment Models in 3D Geomatics

and Computational Geophysics: With

MATLAB Examples Bashar Alsadik
Visit to download the full and correct content document:
https://ebookmass.com/product/adjustment-models-in-3d-geomatics-and-computation
al-geophysics-with-matlab-examples-bashar-alsadik/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Basics of Computational Geophysics Pijush Samui

https://ebookmass.com/product/basics-of-computational-geophysics-
pijush-samui/

Elementary Vector Calculus and Its Applications with

MATLAB Programming (River Publishers Series in
Mathematical, Statistical and Computational Modelling
for Engineering) 1st Edition Nita H. Shah
https://ebookmass.com/product/elementary-vector-calculus-and-its-
applications-with-matlab-programming-river-publishers-series-in-
mathematical-statistical-and-computational-modelling-for-
engineering-1st-edition-nita-h-shah/

Modern Information Optics with MATLAB Yaping Zhang

https://ebookmass.com/product/modern-information-optics-with-
matlab-yaping-zhang/

3D Lung Models for Regenerating Lung Tissue Gunilla

Westergren-Thorsson

https://ebookmass.com/product/3d-lung-models-for-regenerating-
lung-tissue-gunilla-westergren-thorsson/
Computational Frameworks. Systems, Models and
Applications 1st Edition Edition Mamadou Kaba Traore
(Eds.)

https://ebookmass.com/product/computational-frameworks-systems-
models-and-applications-1st-edition-edition-mamadou-kaba-traore-
eds/

Computational Models of Reading: A Handbook 1st Edition

Erik D. Reichle

https://ebookmass.com/product/computational-models-of-reading-a-
handbook-1st-edition-erik-d-reichle/

Computational Geo-electromagnetics: Methods, Models,

and Forecasts: Volume 5 1st Edition Viacheslav V.
Spichak

https://ebookmass.com/product/computational-geo-electromagnetics-
methods-models-and-forecasts-volume-5-1st-edition-viacheslav-v-
spichak/

Distributed Acoustic Sensing in Geophysics: Methods and

Applications Yingping Li

https://ebookmass.com/product/distributed-acoustic-sensing-in-
geophysics-methods-and-applications-yingping-li/

Fundamental Chemistry with MATLAB 1st Edition Daniele

Mazza

https://ebookmass.com/product/fundamental-chemistry-with-
matlab-1st-edition-daniele-mazza/
ADJUSTMENT MODELS IN 3D GEOMATICS
AND COMPUTATIONAL GEOPHYSICS
 Computational Geophysics

ADJUSTMENT
MODELS IN 3D
GEOMATICS AND
COMPUTATIONAL
GEOPHYSICS
With MATLAB Examples

VOLUME 4
BASHAR ALSADIK
Faculty member at Baghdad University – College of Engineering – Iraq (1999–2014)
Research assistant at Twente University – ITC faculty – The Netherlands (2010–2014)
Member of the International Society for Photogrammetry and Remote Sensing ISPRS
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States

# 2019 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopying, recording, or any information storage and retrieval system, without
permission in writing from the publisher. Details on how to seek permission, further information about the
Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance
Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher
(other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our
understanding, changes in research methods, professional practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using
any information, methods, compounds, or experiments described herein. In using such information or methods
they should be mindful of their own safety and the safety of others, including parties for whom they have a
professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any
liability for any injury and/or damage to persons or property as a matter of products liability, negligence or
otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the
material herein.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
ISBN: 978-0-12-817588-0

For information on all Elsevier publications

visit our website at https://www.elsevier.com/books-and-journals

Publisher: Candice Janco

Acquisitions Editor: Amy Shapiro
Editorial project manager: Hilary Carr
Production project manager: Paul Prasad Chandramohan
Cover Designer: Greg Harris
Typeset by SPi Global, India
 Preface
The rapid development in different areas
of science and technology in the last decades
transferred engineering applications into a
new era where the major areas of designa-
tion, analysis, prototyping, and implementa-
tion are being applied in computerized or
automated ways. This means that a large
number of computations that was difficult
or computationally expensive to perform in
the past are today easily handled and signif-
icantly automated or programmed.
An important division of engineering,
namely, traditional surveying and/or geo-
detic engineering, is positively affected by
the developments or advents of satellite nav-
igation, digital mapping, information tech-
nology, robotics, remote sensing, computer
vision, sensors fusion, and other fields related
to Geo-wise applications.
Accordingly:
– Paper-based cartographic plotting has
changed to soft-copy or digital mapping,
web mapping, and geographic
information systems (GIS).
– Traditional photogrammetry is affected
by computer or machine vision techniques
and digital image processing advents.
Further, satellite remote sensing with
different image resolutions and/or
spectral are replacing traditional aerial
photography in many applications.
– Field surveying equipment are
developing away from traditional
theodolites and electronic distance
measurements (EDM) toward robotic total
stations and global navigation satellite
vii
system (GNSS)-based positioning, as an
example.
Hence, the terminology of surveying is
changing in many educational, academic,
and governmental institutions to geomatics,
geoinfomatics, or geospatial engineering,
where three-dimensional observations are
more applied.
Consequently, the author came to the idea
of writing this book that focuses on the ad-
justment of 3D geomatical observations,
which inspires traditional techniques and
puts them in a modern form supported by
solved examples. Further, the book is fo-
cused on presenting both theoretical and
practical context to ensure a clear under-
standing of adjustment computations topic
to the readers. Many numerical examples as
mentioned are introduced and supported by
MATLAB codes to fully understand the topics.
The author should mention that he is not
a professional programmer, and he developed
the codes mainly for educational purposes.
The book is intended for students, lecturers,
and researchers in the field of geomatics and
its divisions. The readers can find the full
list of the code examples published in this link
https://nl.mathworks.com/matlabcentral/
fileexchange/70550-adjustment-models-in-
3d-geomatics.
The structure of the book is designed in 12
chapters to cover essential preliminary and
advanced adjustment topics. Chapter 1 intro-
duces statistical definitions, concepts, and
derivations. The most important method of
the adjustment using the least squares
viii PREFACE
principle and its mathematical derivation is
also introduced. Then, in Chapter 2, error
propagation technique is presented and the
variance-covariance matrix of observations
and unknowns is described. Similarly, the
preanalysis procedure is presented. In
Chapter 3, the adjustment using the condi-
tion equations and the observation equations
is introduced and, at the end, the concept of
the homogeneous least squares method is
presented. For a better understanding, dif-
ferent adjustment models are shown in
Chapter 4 such as intersection and resection
either in 2D or 3D using observed distances,
angles, azimuths, difference in heights, or
computations from images. At the end of
Chapter 4, an important computational geo-
physics application is shown of earthquake
location determination. The general adjust-
ment approach is presented in Chapter 5
when there is more than one observation in
the mathematical model related to several un-
knowns. The important topic of adjustment
with constraints is introduced in Chapter 6.
Three main topics are presented in this chap-
ter: adjustment with constraints, adjustment
with additional parameters, and the adjust-
ment with inner constraints (free nets).
More advanced topics are presented in the
second half of the book, such as the unified
approach of least squares adjustment of
Chapter 7. This is an advanced adjustment
technique where the unknowns have uncer-
tainties and are then processed as observa-
tions in the model.
Chapters 8 and 9 present more related ap-
plications in geomatics, namely the topic of
fitting 3D geometric primitives and the 3D
transformation computations respectively.
Then the book continues to give introduction
to other advanced topics such as the Kalman
filter in Chapter 10 and the nonlinear least
squares using Levenberg-Marquardt in
Chapter 11. Finally, in the Chapter 12 of
the book, the detection, identification, and
adaptation (DIA) of the postadjustment
techniques is introduced. The chapter pre-
sents the blunder detection methods of
data snooping, robust estimation, and the
random sample consensus (RANSAC). In
the book’s Appendix, a MATLAB code is
given for the adjustment of horizontal
geodetic networks.
Bashar Alsadik
The Netherlands
 C H A P T E R

1
Statistical Introduction

1.1 INTRODUCTION

In geomatics, different kinds of measurements (termed as observations in the book) are

applied. The imperfections in instrumentation, weather conditions, and limitations of the
operator0 s skills produce field observations that encompass different kinds of errors. Accord-
ingly, understanding the adjustment of observations and theory of errors are essential to
process and solve geomatical problems. Because we need confidence in applied engineering
or a scientific project, statistical measures of accuracy, precision, and reliability should be
computed and/or standardized. It should be noted that fulfilling the required accuracy stan-
dards will take more time to implement, more professional labor, advanced instrumentation,
and more computing power.
To give real life examples about the importance of this chapter’s perspective and topics
related to adjustment of observations, we list two examples in Fig. 1.1. The first example,
shown in Fig. 1.1A, is about an engineering construction project. Several questions come
to mind: Which end of the bridge is the correctly positioned/aligned one? How accurate it
is? And what error type might be occurred in the calculations and executions?
The second example, shown in Fig. 1.1B, shows racing athletes at the finish line where the
time difference between them is within fractions of a second. So which of the two competitors
is the winner? How accurate is this measuring image system? Is the camera calibrated and
well positioned?

Adjustment Models in 3D Geomatics and Computational Geophysics 1 # 2019 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/B978-0-12-817588-0.00001-5
2 1. STATISTICAL INTRODUCTION

(A) (B)
FIG. 1.1 (A) Which end of the bridge is the correctly aligned one? (B) Which athlete is the gold medal winner?

All of these questions asked of Fig. 1.1 can be answered when we understand different con-
cepts and indices in adjustment of observations and theory of errors such as: accuracy, pre-
cision, reliability, calibration, standard deviation, weighted mean, residual error, most
probable value (MPV), redundancy, etc.

1.2 STATISTICAL DEFINITIONS AND TERMINOLOGIES

• Errors are the differences between observed values and their true values. An error is what
causes values to differ when a measurement is repeated, and none of the results can be
preferred over the others. Although it is not possible to entirely eliminate error in a
measurement, it can be controlled and characterized. We summarized the terminologies
related to errors as follows:
• Gross errors: These are large errors (blunders, mistakes, or outliers) that can be avoided in
the observations; however, they don0 t follow a mathematical or physical model and may
be large or small, positive or negative. With developments in instrumentation and
automated procedures, the main source of gross errors is human-related, for example,
recording and reading errors, or observing a different target than the intended one.
Careful reading and recording of the data can significantly reduce gross errors. It should be
noted that some references don0 t count mistakes as an error type [1].
• Systematic errors
Systematic errors (or bias when having many observations) occur when following
some physical models and therefore can be checked. Systematic errors are either positive or
negative; using proper measuring procedures can eliminate some of them, whereas
some are corrected by using mathematical methods. Systematic error sources are
recognizable and can be reduced to a great extent by careful designation of the observation
system and the selection of its components. In practice, the process of calibration of
instruments, such as cameras in photogrammetry, is to detect and remove systematic
errors.
• Fig. 1.2 illustrates a systematic error-free observation (red curve) biased in a certain
direction and amount (the blue curve).
 1.2 STATISTICAL DEFINITIONS AND TERMINOLOGIES 3

FIG. 1.2 Systematic error (bias) illustration.

• Random errors Random or accidental errors are unavoidable errors that represent residuals
after removing all other types of errors. This type of error occurs in observations because
of limitations in the measuring instruments or due to limitations related to the operator,
among other affecting undetermined factors. It should be noted that random errors
can be positive or negative, and they don0 t follow a physical model. Therefore, they are
processed statistically using the probability theorem because the majority follows normal
distribution. In repetitive observations, the average or mean can be used as the MPV.
As shown in Fig. 1.3, greater random errors cause a greater dispersion of normally
distributed observations around the mean. The dispersion is measured by the standard
deviations at a certain probability as will be shown in Eq. (1.6).

The normal distribution

of X with random errors

The normal distribution of

X without random errors
Frequency

Mean

FIG. 1.3 Observations having a normal probability distribution. (A) With random errors. (B) Without random
errors.

• Uncertainty
All observations have uncertainties, which is a range of values in which the true obser-
vation could lie. An uncertainty estimate should address both systematic and random er-
rors, and therefore is considered the most proper measure of expressing accuracy.
However, in many geomatics problems, the systematic error is disregarded, and only ran-
dom error is included in the uncertainty observation. When only random error is counted
in the uncertainty evaluation, it is an expression of the precision of the observation.
4 1. STATISTICAL INTRODUCTION

• Accuracy and Precision

Accuracy is the closeness between an observed value and its true value (trueness). When
the observations follow normal probability distribution, accuracy can be illustrated as
shown in Fig. 1.4. Therefore, removing systematic errors improves accuracy.
Precision is the closeness between independent observations of a quantity under the same
conditions (Fig. 1.4). It is a measure of how well a value is observed without reference to a
theoretical or true value. The number of divisions on the scale of the measuring device gen-
erally affects the consistency of repeated observations and, therefore, the precision. Because
precision is not based on a true value, there is no bias or systematic error in the value, but
rather it depends solely on the distribution of random errors. The precision of a measure-
ment is usually indicated by the uncertainty or fractional relative uncertainty of a value.
Another useful illustration about the concept of precision and accuracy is shown in Fig. 1.5
using an archery target field in three cases.

Accuracy
related to
systematic error
Probability

True value

Observed value

X
Precision
(related to random error)
FIG. 1.4 Accuracy and precision.

Accurate but less Precise Precise but Inaccurate

because of random errors because of systematic bias
Precise and Accurate

FIG. 1.5 Relation between accuracy and precision.

 1.2 STATISTICAL DEFINITIONS AND TERMINOLOGIES 5
Therefore:
– Reducing systematic errors improves accuracy.
– Reducing random errors improves precision.
An incorrect term is saying that you can reduce random errors by choosing a more accurate
measuring device. The correct term is to say a more precise measuring device. For example, smaller
scale divisions mean a smaller spread, which leads to higher precision whereas a more accu-
rate device would be one that reads true values.
• Reliability
Reliability is an important term in adjustment computations. Reliability refers to the de-
gree of consistency, or reproducibility, of observations. In other words, reliability defines
by how much the adjusted observations must agree with reality. It should be noted that
errors in observations can result in poor reliability. Fig. 1.6 shows examples of reliable
and unreliable observations. The three lines AB, CD, and EF intersect at point P; however,
the left plot shows less reliable observations whereas the right plot shows reliable obser-
vations where the three lines intersect almost exactly at P.

F D F D
B
B P
P

A A

C E C E

FIG. 1.6 (Left) less reliable observations; (right) reliable observations.

• Residual error
Residual error v means the difference between the true value and its observed value.
However, because the true value is impossible to reach, it is statistically compensated
by the MPV. Therefore:

v ¼ MPV  observed value (1.1)

It should be noted that the MPV value of repetitive observations is the arithmetic mean,
whereas for other observations, it is the optimal value that offers the minimal of the squared
residual errors. This concept will be explained in the least squares principle of Section 1.7. It is
important to note that the MPV in the concept of the least squares represents the adjusted
values of observations. For nonlinear geomatical problems, the adjusted values are computed
by running the solution with initial values. Accordingly:

MPV ¼ adjusted values X ¼ approximate values + corrections (1.2)

6 1. STATISTICAL INTRODUCTION

1.3 STATISTICAL INDEXES

• Arithmetic mean x: when observing a quantity x for n times under the same conditions, the
mean x is computed as follows:
Xn
xi
i¼1
x¼ ðn ! ∞ Þ (1.3)
n

• Variance σ 2: the theoretical mean of the squared residual errors that can be computed as
follows:
X n X
n
ð xi  xÞ 2 ð vi Þ 2
i¼1 i¼1
σ2 ¼ ¼ (1.4)
n1 n1
The denominator n  1 is called the redundancy r or the degree of freedom. The minimum
observations we need to determine the variance is only 1 value, which is termed as no; accord-
ingly, the rest of the observations are redundant, and r can be formulated as:
r ¼ n  no (1.5)

• Standard deviation σ: the square root of the variance that expresses by how much the
observations differ from the MPV or the mean value. Further, it is a measure of how spread
out the observations are. Therefore, standard deviation is an expression of precision
computed as:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uX n uX n
u u
u ð xi  xÞ 2 u ð vi Þ 2
t t
i¼1 i¼1
σ¼ ¼ (1.6)
n1 n1
When the reference values are available, the standard deviation is termed as the Root Mean
Squared Error (RMSE), which indicates by how much the observed or derived quantities de-
viate from the reference (true) values.
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uX n
u
u ðxi  xREFRENCE Þ2
t
RMSE ¼ i¼1 (1.7)
n

• Standard error σx : represents the standard deviation of the mean, which is computed as:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uX uX
u n u n
u ð x  x Þ 2 u ð vi Þ 2
u i u
σ t i¼1 t i¼1
σ x ¼ pffiffiffi ¼ ¼ (1.8)
n nðn  1Þ nðn  1Þ
 1.4 NORMAL DISTRIBUTION CURVE 7

EXAMPLE 1.1
Given
A base line AB is observed in meters 10 times as follows:

26.342 26.349 26.351 26.345 26.348

26.350 26.348 26.352 26.345 26.348

Required
Find the MPV, the standard deviation, and standard error of line AB to the nearest mm.

Solution
The MPV for repetitive observations is the mean:
X
n
xi
i¼1
x¼ ¼ 26:348 m
n
The standard deviation is computed as:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uX n
u
u ðxi  xÞ2
t
i¼1
σ¼ ¼ 3 mm
n1
The standard error is computed as:
σ
σ x ¼ pffiffiffi ¼ 1 mm
n
Then
AB ¼ 26:348 m  1 mm

1.4 NORMAL DISTRIBUTION CURVE

In literature related to geomatics, it is presented that all observations taken by surveyors,

such as observed distances and angles, or observations applied by photogrammetrists and
geodesists comply with or follow probability lows. Further, random errors that exist in the
observations are normally distributed (Example 1.2). Accordingly, statistical techniques
can be applied in postprocessing and analysis of the observations.
8 1. STATISTICAL INTRODUCTION

The normal distribution curve or the gaussian curve can be calculated and then plotted
using the following Eq. (1.9).
1 v2
y ¼ pffiffiffiffiffi e 2σ 2 ¼ Keh v
2 2
(1.9)
σ 2π
where
y: the normal probability,
K ¼ σ pnIffiffiffiffi
2π
,
h ¼ 2σ2 called the accuracy index,
2 1

v: the residual error,

σ: the standard deviation,
I: any selected interval of the residual errors,
e: exponential function.

EXAMPLE 1.2
Given
An angle is observed 50 times in degrees as illustrated in Table 1.1:

TABLE 1.1 Measured Angle

40.3429370 40.3414908 40.3373390 40.3363269 40.3400577 40.3441110

40.3414915 40.3373237 40.3352349 40.3398773 40.3416594 40.3419747
40.3366620 40.3385154 40.3412115 40.3433579 40.3452412 40.3384449

40.3365439 40.3381179 40.3401685 40.3397429 40.3377490 40.3466756

40.3424222 40.3434787 40.3401892 40.3456270 40.3394161 40.3384703
40.3382263 40.3422242 40.3374289 40.3401997 40.3402152 40.3411796
40.3432062 40.3405390 40.3419665 40.3393154 40.3409316 40.3441452
40.3368495 40.3435766 40.3413856 40.3364398 40.3411947 40.3443333

40.3371976 40.3395228

Required
– Compute the MPV of the angle and its standard deviation.
– Compute and plot the normal probability distribution curve of residual errors.

Solution
The MPV of the angle is simply the arithmetic mean, therefore:
MPV ¼ mean of angles ¼ 40.3404387 degrees
 1.4 NORMAL DISTRIBUTION CURVE 9
whereas the standards deviation is computed as:
sffiffiffiffiffiffiffiffiffiffiffi
X ffi
v2
σ¼ ¼ 9:83500
n1
To compute the probability y of the normal distribution curve, the following elements are
computed: nI
K ¼ pffiffiffiffiffi ¼ 4:553
σ 2π
1
h2 ¼ ¼ 0:0051688
2σ 2
The computed residuals are grouped between two maximum/minimum limits of 22.4500 . Then,
to have 20 regular intervals for the plotting issue, interval I is computed to be 2.24500 as illustrated in
the following MATLAB code.
Table 1.2. illustrates the details of the computations to relate the residual errors in seconds to the
computed probability y of the normal distribution curve.

TABLE 1.2 Computed Probability of the Measured

Angle
.
K
eh v
2 2
v2 Kv2 y¼ eh v
2 2
v
22.45 504.00 2.61 13.60 0.33
20.20 408.04 2.11 8.25 0.55
17.96 322.56 1.67 5.31 0.86
15.72 247.12 1.28 3.60 1.26

13.47 181.44 0.94 2.56 1.78

11.22 125.89 0.65 1.92 2.37
8.98 80.64 0.42 1.52 3.00
6.74 45.43 0.23 1.26 3.61
4.49 20.16 0.10 1.11 4.10
2.24 5.02 0.03 1.03 4.42

0.00 0.00 0.00 1.00 4.55

2.24 5.02 0.03 1.03 4.42
4.49 20.16 0.10 1.11 4.10
6.74 45.43 0.23 1.26 3.61
8.98 80.64 0.42 1.52 3.00
11.22 125.89 0.65 1.92 2.37

13.47 181.44 0.94 2.56 1.78

15.72 247.12 1.28 3.60 1.26
17.96 322.56 1.67 5.31 0.86
20.20 408.04 2.11 8.25 0.55
22.45 504.00 2.61 13.60 0.33
10 1. STATISTICAL INTRODUCTION

Then the normal distribution curve and histogram can be plotted as shown in Fig. 1.7.

4.5

3.5

3
Probability

2.5

1.5

0.5

0
–25 –20 –15 –10 –5 0 5 10 15 20 25
Residuals sec.
FIG. 1.7 The normal distribution of residuals in seconds.

MATLAB code
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%% chapter 1 - example 1.2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc,clear, close all
%%%%%%%%%%%%% create normally distributed observed angles %%%%%%%%%%%%%%%%%
ang=40.34+10*(randn(50,1)/3600); %% adding random normal noise of 10 sec.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% compute residuals
resid =3600*(ang-mean(ang));
Resid=unique(round(resid*100)/100);
m=max(abs(Resid)); %% make equal intervals
I= ((2*m)/20) % 20 intervals % can be changed by the user
v=-m:I:m; v=round(v’*100)/100; % sample the residuals equally in seconds

v2=round(v.^2*100)/100; % squared residuals

sigma=std(ang);sigma=sigma*3600; % standard deviation in seconds
n=size(ang,1); % total number of observations
 1.5 CUMULATIVE DISTRIBUTION FUNCTION 11
k1=(n*I)/(sigma*sqrt(2*pi)); % k1 of the normal probab. distribution
k2=1/(2*sigma^2); % k2 of the normal probab. distribution
k2v2=round(k2*v2*100)/100;
exp_k2v2=round(exp(k2v2)*100)/100;
y=round((k1./exp_k2v2)*100)/100; % Y probability
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
plot(v,y,’-’,’linewidth’,3); hold on; % plot the normal curve
xlabel(’RESIDUALS sec.’)
ylabel(’PROBABILITY’)
grid on
disp(’SUMMARIZED CALCULATIONS’)
disp(’-————————————————————’)
T = table(v, v2, k2v2, exp_k2v2, y)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
bar(v,y);alpha(.5); % plot the bars
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1.5 CUMULATIVE DISTRIBUTION FUNCTION

A set of observations that are normally distributed can be represented as a histogram or

distribution curve as shown in Example 1.2.
On the other hand, the Cumulative Distribution Function (CDF) shows the percentage or
relative count of the sorted observation values over the observations themselves [2]. This
is, in fact, the integral of the normal distribution histogram.
Let’s assume we have 100 observations of an angle where the mean is 50 degrees with a
precision of 10 degrees. We can represent the observed values either in a histogram or in
a CDF plot as follows:
MATLAB code
% angle x biased randomly
x=randn(100,1)*10+50
subplot (1,2,1)
hist(x)
subplot(1,2,2)
cdfplot(x)

The CDF plot in Fig. 1.8B explains that 90% of the observed angles are less than or equal to
60 degrees. This property is easy to interpret even for a nonprofessional. Therefore, key values
such as minimum, maximum, median, percentiles, etc. can be directly read from the diagram.
Accordingly, CDF is an efficient description for random variable uncertainty and is a useful
tool for comparing the distribution of different sets of data.
Let’s assume that three observers A, B, and C measured the same angle 300 times each. To
compare the skill of the three observers in observing the angle, we can use the CDF plot as
shown in Fig. 1.9.
12 1. STATISTICAL INTRODUCTION

FIG. 1.8 (A) Normal distribution histogram, (B) The corresponding CDF plot.

FIG. 1.9 The CDF plots of the angle observed by three surveyors.

MATLAB code
clear;clc; close all
% angle biased randomly
A=randn(300,1)*10+50
hA=cdfplot(A);hold on
set( hA(:,1), ’Linewidth’, 3);
B=randn(300,1)*10+48
 1.6 THE PROBABLE ERROR AND LEVELS OF REJECTION 13
hB=cdfplot(B);hold on
set( hB(:,1), ’Linewidth’, 3);
C=randn(300,1)*10+53
hC=cdfplot(C);hold on
set( hC(:,1), ’Linewidth’, 3);
lgd =legend(’observer A’, ’observer B’, ’observer C’,’fontsize’);
lgd.FontSize = 12;

1.6 THE PROBABLE ERROR AND LEVELS OF REJECTION

The probable error Pe_50% is defined as the 50% or equal probability that a true error can
arise. In other words, 50% probability occurs when j vi j > Pe_50% and 50% when jvi j < Pe_50% as
shown in Fig. 1.10.

FIG. 1.10 The 50% probable error.

For a group of observations, the 50% probable error is the median of residual errors after
sorting them in ascending or descending order, and therefore it is sometimes termed as the
median absolute deviation (MAD) (Chapter 12). However, by using this approach, the probable
error is not sensitive to gross errors as shown in the following illustrations where the probable
error is 500 for both error sets despite the existence of gross error of 9400 .
1st set: 1″ 1″ 3″ 5″ 6″ 7″ 11″
2nd set: 1″ 1″ 3″ 5″ 6″ 7″ 94″
Statistically, to relate the probable error Pe_50% with the standard deviations, we can apply
the following derivation by the integration of the normal distribution curve equation as:
ð +Pe v2
1
Area ¼ pffiffiffiffiffiffiffiffi e 2σ 2 dv ¼ 0:5 (1.10)
2πσ Pe

Accordingly, the probable error is evaluated as:

2
Pe_50% ¼ 0:6745σ ffi σ (1.11)
3
14 1. STATISTICAL INTRODUCTION

where σ is the standard deviation and v is residual error.

Applying the same derivation procedure, we can compute the probable error for any given
standard deviation. Hence, the probable error for the range  σ is found to be:
Pðσ  ε  + σ Þ ¼ 0:683 (1.12)
Therefore, the probable error between  σ and + σ is 68% of the total area of the normal
curve as shown in Fig. 1.11. As an illustration, when a line length is observed with a precision
of 0.06 m, it means there is a probability of 68% that the true error of observations is equal to
or less than 0.06 m.

Inflection point Inflection point

68% of Area

FIG. 1.11 Probable error in the range of σ.

In the same manner, the probable error of the intervals 2σ and 3σ are computed as
shown in Eq. (1.13) and Fig. 1.12.
Pð2σ  ε  2σ Þ ¼ 0:955
(1.13)
Pð3σ  ε  3σ Þ ¼ 0:997

FIG. 1.12 Different standard deviations limits of the normal distribution curve.
Another random document with
no related content on Scribd:
T he Chase I sing, hounds, and their various breed,
And no less various use. O thou, great Prince!
Whom Cambria’s towering hills proclaim their lord,
Deign thou to hear my bold, instructive song.
While grateful citizens, with pompous shew,
Rear the triumphal arch, rich with the exploits
Of thy illustrious house; while virgins pave
Thy way with flowers, and, as the royal youth
Passing they view, admire, and sigh in vain;
While crowded theatres, too fondly proud
Of their exotick minstrels, and shrill pipes,
The price of manhood, hail thee with a song,
And airs soft-warbling; my hoarse-sounding horn

BOOK I THE CHASE v. 14-38.

Invites thee to the chase, the sport of kings;
Image of war, without its guilt. The Muse
Aloft on wing shall soar, conduct with care
Thy foaming courser o’er the steepy rock,
Or, on the river bank, receive thee safe,
Light-bounding o’er the wave, from shore to shore.
Be thou our great protector, gracious youth!
And if, in future times, some envious prince,
Careless of right, and guileful, should invade
Thy Britain’s commerce, or should strive, in vain,
To wrest the balance from thy equal hand,
Thy hunter-train, in cheerful green array’d,
A band undaunted, and innured to toils,
Shall compass thee around, die at thy feet,
Or hew thy passage through the embattled foe,
And clear thy way to fame: inspired by thee,
The nobler chase of glory shall pursue,
Through fire, and smoke, and blood, and fields of death.
Nature, in her productions slow, aspires,
By just degrees, to reach perfection’s highth:
So mimick art works leisurely, till time
Improve the piece, or wise experience give
The proper finishing. When Nimrod bold,
That mighty hunter, first made war on beasts,
And stain’d the woodland green with purple dye,

BOOK I THE CHASE v. 39-63.

New, and unpolish’d, was the huntsman’s art;
No stated rule, his wanton will his guide.
With clubs and stones, rude implements of war,
He arm’d his savage bands, a multitude
Untrain’d: of twining osiers form’d, they pitch
Their artless toils, then range the desert hills,
And scour the plains below: the trembling herd
Start at the unusual sound, and clamorous shout,
Unheard before; surprised, alas! to find
Man now their foe, whom erst they deem’d their lord;
But mild, and gentle, and by whom, as yet,
Secure they grazed. Death stretches o’er the plain,
Wide-wasting, and grim slaughter, red with blood:
Urged on by hunger keen, they wound, they kill;
Their rage, licentious, knows no bound: at last,
Incumber’d with their spoils, joyful they bear,
Upon their shoulders broad, the bleeding prey.
Part on their altars smokes a sacrifice
To that all-gracious Power, whose bounteous hand
Supports his wide creation: what remains,
On living coals they broil, inelegant
Of taste, nor skill’d, as yet, in nicer arts
Of pamper’d luxury. Devotion pure,
And strong necessity, thus first began
The chase of beasts; though bloody was the deed,

BOOK I THE CHASE v. 64-88.

Yet without guilt: for the green herb, alone,
Unequal to sustain man’s labouring race,
Now every moving thing that lived on earth,
Was granted him for food. So just is Heaven,
To give us in proportion to our wants.
Or chance, or industry, in after-times,
Some few improvements made; but short, as yet,
Of due perfection. In this isle, remote,
Our painted ancestors were slow to learn,
To arms devote, of the politer arts
Nor skill’d, nor studious; till, from Neustria’s coasts,
Victorious William to more decent rules
Subdued our Saxon fathers, taught to speak
The proper dialect; with horn and voice
To cheer the busy hound, whose well-known cry
His listening peers approve with joint acclaim.
From him successive huntsmen learn’d to join,
In bloody social leagues, the multitude
Dispersed, to size, to sort their various tribes,
To rear, feed, hunt, and discipline the pack.
Hail, happy Britain! highly favour’d isle,
And Heaven’s peculiar care; to thee ’tis given
To train the sprightly steed, more fleet than those
Begot by winds, or the celestial breed
That bore the great Pelides through the press

BOOK I THE CHASE v. 89-113.

Of heroes arm’d, and broke their crowded ranks;
Which, proudly neighing, with the sun begins
Cheerful his course; and ere his beams decline,
Has measured half thy surface unfatigued.
In thee alone, fair land of liberty!
Is bred the perfect hound, in scent and speed
As yet unrivall’d; while in other climes
Their virtue fails, a weak degenerate race.
In vain malignant steams, and winter fogs,
Load the dull air, and hover round our coasts;
The huntsman, ever gay, robust, and bold,
Defies the noxious vapour, and confides
In this delightful exercise, to raise
His drooping head, and cheer his heart with joy.
Ye vigorous youths, by smiling fortune bless’d
With large demesnes, hereditary wealth,
Heap’d copious by your wise forefathers’ care,
Hear, and attend; while I the means reveal
To enjoy those pleasures, for the weak too strong,
Too costly for the poor: to rein the steed
Swift-stretching o’er the plain, to cheer the pack,
Opening in concerts of harmonious joy,
But breathing death. What though the gripe severe
Of brazen-fisted time, and slow disease
Creeping through every vein, and nerve unstrung,

BOOK I THE CHASE v. 114-138.

Afflict my shatter’d frame, undaunted still,
Fix’d as the mountain ash, that braves the bolts
Of angry Jove, though blasted, yet unfall’n;
Still can my soul, in fancy’s mirrour, view
Deeds glorious once, recall the joyous scene
In all its splendours deck’d, o’er the full bowl
Recount my triumphs pass’d, urge others on
With hand and voice, and point the winding way:
Pleased with that social sweet garrulity,
The poor disbanded veteran’s sole delight.
First, let the kennel be the huntsman’s care;
Upon some little eminence erect,
And fronting to the ruddy dawn; its courts
On either hand wide opening to receive
The sun’s all-cheering beams, when mild he shines,
And gilds the mountain tops. For much the pack
(Roused from their dark alcoves) delight to stretch
And bask in his invigorating ray:
Warn’d by the streaming light, and merry lark,
Forth rush the jolly clan; with tuneful throats
They carol loud, and, in grand chorus join’d,
Salute the new-born day. For not alone
The vegetable world, but men and brutes
Own his reviving influence, and joy
At his approach. Fountain of light! if chance

BOOK I THE CHASE v. 139-163.

Some envious cloud veil thy refulgent brow,
In vain the Muses aid, untouch’d, unstrung,
Lies my mute harp, and thy desponding bard
Sits darkly musing o’er the unfinish’d lay.
Let no Corinthian pillars prop the dome,
A vain expense, on charitable deeds
Better disposed, to clothe the tatter’d wretch
Who shrinks beneath the blast, to feed the poor,
Pinch’d with afflictive want: for use, not state,
Gracefully plain let each apartment rise.
O’er all let cleanliness preside; no scraps
Bestrew the pavement, and no half-pick’d bones,
To kindle fierce debate, or to disgust
That nicer sense, on which the sportsman’s hope,
And all his future triumphs, must depend.
Soon as the growling pack, with eager joy,
Have lapp’d their smoking viands, morn or eve,
From the full cistern lead the ductile streams,
To wash thy court, well-paved; nor spare thy pains,
For much to health will cleanliness avail.
Seek’st thou for hounds to climb the rocky steep,
And brush the entangled covert, whose nice scent
O’er greasy fallows, and frequented roads,
Can pick the dubious way? Banish far off
Each noisome stench, let no offensive smell

BOOK I THE CHASE v. 164-188.

Invade thy wide inclosure, but admit
The nitrous air, and purifying breeze.
Water and shade no less demand thy care:
In a large square the adjacent field inclose;
There plant, in equal ranks, the spreading elm,
Or fragrant lime; most happy thy design,
If, at the bottom of thy spacious court,
A large canal, fed by the crystal brook,
From its transparent bosom shall reflect
Downward thy structure and inverted grove.
Here, when the sun’s too potent gleams annoy
The crowded kennel, and the drooping pack,
Restless and faint, loll their unmoisten’d tongues,
And drop their feeble tails, to cooler shades
Lead forth the panting tribe; soon shalt thou find
The cordial breeze their fainting hearts revive:
Tumultuous soon they plunge into the stream,
There lave their reeking sides, with greedy joy
Gulp down the flying wave; this way and that,
From shore to shore, they swim, while clamour loud,
And wild uproar, torments the troubled flood;
Then on the sunny bank they roll and stretch
Their dripping limbs; or else in wanton rings
Coursing around, pursuing and pursued,
The merry multitude disporting play.

BOOK I THE CHASE v. 189-213.

 But here, with watchful and observant eye,
Attend their frolicks, which too often end
In bloody broils and death. High o’er thy head
Wave thy resounding whip, and, with a voice
Fierce-menacing, o’er-rule the stern debate,
And quench their kindling rage; for oft, in sport
Begun, combat ensues; growling they snarl,
Then on their haunches rear’d, rampant they seize
Each other’s throats, with teeth and claws, in gore
Besmear’d, they wound, they tear, till on the ground,
Panting, half dead, the conquering champion lies:
Then sudden all the base ignoble crowd,
Loud-clamouring, seize the helpless worried wretch,
And, thirsting for his blood, drag different ways
His mangled carcase on the ensanguined plain.
O breasts of pity void! to oppress the weak,
To point your vengeance at the friendless head,
And, with one mutual cry, insult the fall’n!
Emblem too just of man’s degenerate race.
Others apart, by native instinct led,
Knowing instructor! ’mong the ranker grass
Cull each salubrious plant, with bitter juice
Concoctive stored, and potent to allay
Each vicious ferment. Thus the hand divine
Of Providence, beneficent and kind

BOOK I THE CHASE v. 213-236.

To all his creatures, for the brutes prescribes
A ready remedy, and is himself
Their great physician! Now grown stiff with age,
And many a painful chase, the wise old hound,
Regardless of the frolick pack, attends
His master’s side, or slumbers, at his ease,
Beneath the bending shade; there, many a ring
Runs o’er in dreams; now on the doubtful foil
Puzzles perplex’d, or doubles intricate
Cautious unfolds; then, wing’d with all his speed,
Bounds o’er the lawn to seize his panting prey,
And in imperfect whimp’ring speaks his joy.
A different hound, for every diff’rent chase,
Select with judgment; nor the timorous hare
O’ermatch’d destroy, but leave that vile offence
To the mean, murd’rous, coursing crew, intent
On blood and spoil. Oh blast their hopes, just Heaven!
And all their painful drudgeries repay
With disappointment, and severe remorse.
But husband thou thy pleasures, and give scope
To all her subtle play: by nature led,
A thousand shifts she tries; to unravel these
The industrious beagle twists his waving tail,
Through all her labyrinths pursues, and rings

BOOK I THE CHASE v. 237-260.

Her doleful knell. See there, with countenance blithe,
And with a courtly grin, the fawning hound
Salutes thee, cowering, his wide opening nose
Upward he curls, and his large sloe-black eyes
Melt in soft blandishments, and humble joy;
His glossy skin, or yellow-pied, or blue,
In lights or shades by nature’s pencil drawn,
Reflects the various tints; his ears and legs,
Fleckt here and there, in gay enamell’d pride
Rival the speckled pard; his rush-grown tail
O’er his broad back bends in ample arch;
On shoulders clean, upright and firm he stands;
His round cat foot, straight hams, and wide-spread thighs,
And his low-dropping chest, confess his speed,
His strength, his wind, or on the steepy hill,
Or far-extended plain; in every part
So well proportion’d, that the nicer skill
Of Phidias himself can’t blame thy choice.
Of such compose thy pack: but here a mean
Observe; nor the large hound prefer, of size
Gigantick; he in the thick-woven covert
Painfully tugs, or in the thorny brake
Torn and embarrass’d, bleeds: but if too small,
The pigmy brood in every furrow swims;

BOOK I THE CHASE v. 261-285.

Moil’d in the clogging clay, panting they lag
Behind inglorious; or else shivering they creep,
Benumb’d and faint, beneath the shelt’ring thorn.
For hounds of middle size, active and strong,
Will better answer all thy various ends,
And crown thy pleasing labours with success.
As some brave captain, curious and exact,
By his fix’d standard forms, in equal ranks,
His gay battalion, as one man they move,
Step after step, their size the same, their arms
Far gleaming, dart the same united blaze:
Reviewing generals his merit own;
How regular! how just! and all his cares
Are well repaid, if mighty George approve.
So model thou thy pack, if honour touch
Thy generous soul, and the world’s just applause.
But above all take heed, nor mix thy hounds
Of diff’rent kinds; discordant sounds shall grate
Thy ears offended, and a lagging line
Of babbling curs disgrace thy broken pack.
But if the amphibious otter be thy chase,
Or stately stag, that o’er the woodland reigns;
Or if the harmonious thunder of the field
Delight thy ravish’d ears; the deep-flew’d hound
Breed up with care, strong, heavy, slow, but sure,

BOOK I THE CHASE v. 286-310.

Whose ears, down-hanging from his thick round head,
Shall sweep the morning dew; whose clanging voice
Awake the mountain echo in her cell,
And shake the forests: the bold Talbot kind
Of these the prime, as white as Alpine snows;
And great their use of old. Upon the banks
Of Tweed, slow-winding through the vale, the seat
Of war and rapine once, ere Britons knew
The sweets of peace, or Anna’s dread commands
To lasting leagues the haughty rivals awed,
There dwelt a pilfering race; well train’d and skill’d
In all the mysteries of theft, the spoil
Their only substance, feuds and war their sport:
Not more expert in every fraudful art
The arch felon was of old, who by the tail
Drew back his lowing prize: in vain his wiles,
In vain the shelter of the covering rock,
In vain the sooty cloud, and ruddy flames,
That issued from his mouth; for soon he paid
His forfeit life; a debt how justly due
To wrong’d Alcides, and avenging Heaven!
Veil’d in the shades of night, they ford the stream,
Then prowling far and near, whate’er they seize
Becomes their prey; nor flocks nor herds are safe,
Nor stalls protect the steer, nor strong-barr’d doors

BOOK I THE CHASE v. 311-335.

Secure the favourite horse. Soon as the morn
Reveals his wrongs, with ghastly visage wan,
The plunder’d owner stands, and from his lips
A thousand thronging curses burst their way:
He calls his stout allies, and in a line
His faithful hound he leads; then, with a voice
That utters loud his rage, attentive cheers:
Soon the sagacious brute, his curling tail
Flourish’d in air, low-bending plies around
His busy nose, the steaming vapour snuffs
Inquisitive, nor leaves one turf untried;
Till, conscious of the recent stains, his heart
Beats quick; his snuffling nose, his active tail,
Attest his joy; then, with deep-opening mouth,
That makes the welkin tremble, he proclaims
The audacious felon; foot by foot he marks
His winding way, while all the listening crowd
Applaud his reasonings. O’er the watery ford,
Dry sandy heaths, and stony barren hills,
O’er beaten paths, with men and beasts distain’d,
Unerring he pursues; till at the cot
Arrived, and seizing by his guilty throat
The caitiff vile, redeems the captive prey:
So exquisitely delicate his sense!
Should some more curious sportsman here inquire,

BOOK I THE CHASE v. 336-360.

Whence this sagacity, this wond’rous power,
Of tracing step by step, or man or brute;
What guide invisible points out their way,
O’er the dank marsh, bleak hill, and sandy plain?
The courteous Muse shall the dark cause reveal.
The blood that from the heart incessant rolls
In many a crimson tide, then here and there,
In smaller rills disparted, as it flows,
Propell’d, the serous particles evade
Through the open pores, and, with the ambient air
Entangling, mix: as fuming vapours rise,
And hang upon the gently-purling brook,
There by the incumbent atmosphere compress’d.
The panting chase grows warmer as he flies,
And through the net-work of the skin perspires;
Leaves a long streaming trail behind, which, by
The cooler air condensed, remains, unless
By some rude storm dispersed, or rarefied
By the meridian sun’s intenser heat:
To every shrub the warm effluvia cling,
Hang on the grass, impregnate earth and skies:
With nostrils opening wide, o’er hill, o’er dale,
The vigorous hounds pursue, with every breath
Inhale their grateful steam; quick pleasures sting
Their tingling nerves, while they their thanks repay,

BOOK I THE CHASE v. 361-385.

And in triumphant melody confess
The titillating joy. Thus on the air
Depend the hunter’s hopes. When ruddy streaks
At eve, forebode a blust’ring stormy day,
Or lowering clouds blacken the mountain’s brow;
When nipping frosts, and the keen biting blasts
Of the dry parching east menace the trees,
With tender blossoms teeming, kindly spare
Thy sleeping pack, in their warm beds of straw
Low-sinking, at their ease; listless they shrink
Into some dark recess, nor hear thy voice,
Though oft invoked; or, haply, if thy call
Rouse up the slumbering tribe, with heavy eyes,
Glazed, lifeless, dull, downward they drop their tails
Inverted; high on their bent backs erect
Their pointed bristles stare; or ’mong the tufts
Of ranker weeds, each stomach-healing plant
Curious they crop, sick, spiritless, forlorn.
These inauspicious days, on other cares
Employ thy precious hours; the improving friend
With open arms embrace, and from his lips
Glean science, season’d with good-natured wit.
But if the inclement skies and angry Jove
Forbid the pleasing intercourse, thy books
Invite thy ready hand, each sacred page

BOOK I THE CHASE v. 386-399.

Rich with the wise remarks of heroes old.
Converse familiar with the illustrious dead;
With great examples of old Greece or Rome
Enlarge thy free-born heart; and bless kind Heaven,
That Britain yet enjoys dear liberty,
That balm of life, that sweetest blessing; cheap,
Though purchased with our blood. Well bred, polite,
Credit thy calling. See! how mean, how low,
The bookless, sauntring youth, proud of the skut
That dignifies his cap, his flourish’d belt,
And rusty couples gingling by his side.
Be thou of other mould; and know, that such
Transporting pleasures, were by Heaven ordain’d
Wisdom’s relief, and Virtue’s great reward.
BOOK II
 Argument.
Of the power of instinct in brutes. Two remarkable instances in the hunting
of the roebuck, and in the hare going to seat in the morning. Of the variety
of seats or forms of the hare, according to the change of the season,
weather, or wind. Description of the hare-hunting in all its parts,
interspersed with rules to be observed by those who follow that chase.
Transition to the Asiatick way of hunting, particularly the magnificent
manner of the Great Mogul, and other Tartarian princes, taken from
Monsieur Bernier, and the History of Gengis Cawn the Great. Concludes
with a short reproof of tyrants and oppressors of mankind.