Image Recognition and Landmark Detection: A Literature Review

This paper conducts a literature review on image recognition and landmark detection, covering topics such as feature extraction, image processing, visual search architectures, and image search optimization. The goal is to create a real-time landmark recognition Android app that can simplify content understanding and geolocation of images returned via GPS.

Abstract
Essentially, computer vision is the computer's ability to gain a high-level understanding
from text, digital images or videos that it is presented with.
Within the arena of machine learning, computer visualisation and image processing,
image recognition presents a broad horizon of challenging tasks.
How to extract optimal, representative key features that can reflect the intrinsic content of
an image as accurately and efficiently as possible remains both a primary interest and an
exacting task within the domain of computer vision.
Keywords: - Image Recognition, Machine Learning, Data-set, Bag-of-Words, GPS
Receiver, Key Point Detection/Description, PICASA, Monulens and Google Goggles.
1

Table of Contents
1.0 Introduction.....................................................................................................................4
2.0 Literature Review: ..........................................................................................................6
2.1 A Survey of Major Image/Landmark Recognition & Comparison Techniques:..............6
2.2 Mobile Visual Search Architectures:...........................................................................11
3.0 Summary...................................................................................................................... 18
4.0 Conclusion....................................................................................................................19
5.0 References................................................................................................................... 20
6.0 Bibliography..................................................................................................................24

3

1.0 Introduction
A touristic landmark is an instantly recognisable building or site (such as a monument or a
cathedral).
As a traveller, gathering real-time, interactive and
relevant information about a landmark or monument and
its surrounding areas of interest are important aspects of
an individual's journey because of the cultural and
historical aspects it can present.
The World Wide Web is readily abundant with images
and video recordings.
Creating a mobile application which has the ability to
recognise and match a vast number of landmarks
efficiently still remains a challenge due to the sheer number of images that are required to
be searched within a database/dataset, along with the presence of possible visual
distortions such as external litter, illuminated amendment and dynamic geometry of the
imaging devices by which the visual media was recorded (Rekhansh et al., 2015).
However, with the vast amount of landmark pictures emerging within the World Wide Web
and the advancement of landmark picture sharing via websites such as Picasa and Flickr,
the requirement for a computer conception to recognise landmarks universally through the
creation of reliable image identification engines and algorithms is necessary (Rohr, 2010).
In this research work, a literature review will be conducted to assess research
developments in the field of image recognition and the discoveries and progress that has
been made towards technologies such as image feature extraction, image processing,
visual search architectures and image search optimisation in general.
Finally, this paper will be summarised, with primary conclusions highlighted and some
potential research directions and techniques that can be further explored and improved
upon identified.
4
Fig.1: Famous Landmarks (Google.com,
2016)

Only through this process may an application be bestowed with image recognition
capabilities advanced enough to fully recognise and understand a diverse range of images
using neural networks (Parker, 1996).
This includes taking into consideration a number of issues that often present themselves
when creating such a device, such as the variations that are guaranteed to emerge in any
given landmark from one observation to the next (Yairi, Hirama, and Hori, 2003).
The ultimate goal of the proposed project is to create a real-time landmark recognition
Android app.
The device will use a large dataset of images which will be stored on the device itself for
matching. By identifying a given landmark, the application will simplify both content
understanding and the geolocation of images returned via GPS, enabling a topographical
representation and navigation of landmarks in the local area to provide appropriate tour
guide suggestions and guidance via online resources such as Yelp and Google Maps.
However, as already highlighted, efficiency is a challenge for such a potentially unstable,
large-scale image recognition system (Pinto, Cox and DiCarlo, 2008).

2.0 Literature Review:
2.1 A Survey of Major Image/Landmark Recognition & Comparison Techniques:
Various works have been carried out within the area of image recognition adhering to the
requirements of the main specifications.
On observing the dearth of methods for the identification and correlation of specific details
from images stored on the World Wide Web and the limits of purely camera-based,
contemporary programs, various researchers have attempted to come up with a means of
landmark recognition and modelling (some on a global-scale).
Colios et al. (2001) experimented with automated landmark identification for robots by
means of both projective and point-permutation invariant vectors. These vectors were
used to identify landmark patterns based on workspace planar features, enabling the
creation of direct, point-to-point correspondences in an indoor setting.
This, in turn, allowed for the use of both projectivity constraints and the convex hull to
identify matches in sets of five different images.
There was a noticeable margin of error; though this was reduced through the use of sub-
landmarks as outlier patterns.
However, this approach was limited to indoor environments only, rendering it's
effectiveness to be restrictive (Rekhansh et al., 2015).
As part of a different project, vision-based landmark recognition was also used by Mata et
al. (2002) to perform topological localisation of mobile robots.
Both natural and artificial landmarks alike were used here, enabling the creation of a
search function whereby pattern identification/recognition techniques were applied to
digital images.
6

Not only did this enable robot navigation, but it also allowed for text strings, when present,
to be captured and understood from within landmark pictures.
Tests on robots, such as a B21 mobile robot, proved the usefulness of this approach.
Unfortunately, the method did not touch on any network or internet-based navigation, so a
potential limitation occurs in the navigation capabilities of robots when utilising this
particular method.
This, in turn, suggests that computer vision still had some way to go before it could
combine with web-based localisation and map construction technologies
(Yairi, Hirama, and Hori, 2003; Rekhansh et al., 2015).
Pablo Sala et al. (2006) presented a unique chart theoretic plan of the issue of
consequently extricating an ideal arrangement of points of interest from an environment for
visual navigational use.
It's obstinate intricacy propelled the requirement for approximation algorithms, and
displayed six algorithmic calculations.
To assess them, the authors initially tried them on a simulator, where they could fluctuate
the state of the environment, the number and state of objects, the dissemination of the
elements and the perceivability of the components.
The calculation that accomplished the best outcomes on engineered information was then
displayed through genuine visibility data.
The subsequent decompositions revealed expansive areas on the world in which a small
number of components could be tracked to support proficient online localisation.
Their detailing and arrangement of the issues were general, and could suit other classes of
image features.
Elmogy et al. (2009) developed a robust and fast landmark recognition system for
online landmark recognition during robot navigation. Colour histograms of the
landmarks encountered were used to provide a proximate initial estimate of the
landmark.

The resulting hypotheses was then processed to calculate an accurate estimation of
the landmark.
A topological map displaying the routes was used to reduce the processing time by
only processing landmarks mentioned in the route description, ignoring the other
landmarks during navigation.
The robot’s stereo vision was also combined with the classified landmarks for
locating the nearest landmark to the automation and to calculate the landmark’s
geographical position in the real world.
Meyer-Delius et al. (2011) introduced a point of interest situation approach that attempted
to diminish the overall uncertainty within an environment to enhance the navigational
execution of a portable robot.
They proposed a measure for the uniqueness of a robot in light of the appearance of the
earth as seen by the automation.
Because of the combinatorial nature of the landmark arrangement issue, they presented
an estimated approach that incrementally chose historic point areas from an arrangement
of candidate areas, and in this way augmented the normal uniqueness in the environment.
Moreover, they depicted a solid application with regards to limitation with laser run
scanners, given a grid based representation of the environment.
They assessed their approach for various situations in duplicating and utilising genuine
information.
The outcome revealed that their approach yielded generous enhancements in localised
performance with robots.
8

Being primarily concerned with efficiency issues of GPS based web enabled landmark
recognition, Yan-Tao et al. (2009) discovered that while the low-dimensionality of GPS
coordinates does not demand high-end equipment or efficiency, input images in high
quantities may easily inhibit the process, especially when landmark models are particularly
large.
Methods to improve the efficiency of landmark image mining and recognition of query
images were explored through the use of techniques such as parallel computing, efficiency
and hierarchical clustering, and local feature indexing for easier and efficient matching.
These methods produced improvements in the speed and accuracy of the clustering and
image matching process.
However, parallel computing is not something that a mobile device can easily execute,
being a feature that generally shows its worth only in supercomputer clusters
(Almasi and Gottlieb, 1987; Hwang, Fox, and Dongarra, 2011).
Nevertheless, efficiency and hierarchical clustering and local feature indexing is something
that can definitely be implemented efficiently at software-level (Yan-Tao Zheng et al.,
2009).
Indeed, modern tablets and mobile phones have reached a point where they possess
computing and graphics hardware which are on a par with those of the personal
computers of barely two or three decades ago (Girod et al., 2011).
Additionally, with the current ubiquity of 3G and Wi-Fi for mobile devices, it is not out of the
question that a cloud-based feature could be added to an application, to provide it with
remote access to parallel computing resources via the cloud (Grama, Gupta, and Karypis,
2003; Landfeldt, 2009; Hwang, Fox, and Dongarra, 2011).

More recent landmark recognition techniques may be separated into two categories:
1. Template-based representation
2. Feature-based representation (Chan and Baciu, 2012)
The template-based approach to landmark recognition employs holistic texture features
and creates arrays from complete building/location image patterns. These arrays are then
compared by means of metrics such as the Euclidian distance (Ansari and Li, 1993).
The Eigenspace method is currently the most widely accepted approach to template-
based landmark recognition. Based on the PCE technique (Principal Components
Analysis), this method enables high-speed landmark recognition via dimensionality
reduction (Chan and Baciu, 2012).
Feature-based representation, meanwhile, looks at a landmark’s geometrical features,
such as the relative position, or specific elements of, pillars, walls, doors, windows,
crossbeams and decorations. Based on these feature points, equations may be used to
calculate landmark positioning, in a way similar to camera-based positioning
(Seo and Yoo, 2004; Gillner, Weiß, and Mallot, 2008).
However, neither of these approaches is without its limitations.
Template-based methods (Eigenspace and fisher-space) lose most of their usefulness
when trying to parse variations in lighting, though wavelet decomposition may help to
mitigate this weakness somewhat (Chan and Baciu, 2012).
Feature-based applications, like vision-based positioning, demand that an application
constantly compares multiple visual scenes, preferably of high resolution.
10

This, whilst so far successfully applied to robots, demands computing resources that are
generally too large and consuming for a mobile application (Li and Yang, 2003; Li et al.,
2011).

2.2 Mobile Visual Search Architectures:
For about a decade there has been a growing interest in both landmark recognition and
optimisation techniques within the computer vision community.
However, its use in a mobile application setting have been relatively recent.
Mobile image-based retrieval applications emerge with their own set of challenges, one of
which is the consideration of how much of the processing (or which parts) should be
carried out by the mobile device itself, and what should be off-loaded to a server (Girod et
al., 2011; Li and Yap, 2012).
Such a device will require a database that can be searched quickly across many images.
The main consideration here is how to represent and store the database that will be used.
Most commercial, large-scale applications such as Google Goggles rely heavily on server-
side support and processing, and utilises an approach that relies on incorporating priors
from noisy user location data, using the image content itself for identification.
However, the creation of an app that does not rely on any server-side support for
processing would help in curbing the latency, power and network requirements that are
often associated with communication to and from a server when using such a device
(Chen, 2013).
Monulens is an application that does not rely on server side support. The whole procedure
happens on the gadget itself and the match is displayed through the utilisation of a
significant dataset.
While client side executions offer noteworthy advantages, one limitation of a server
autonomous execution is that data-intensive applications devour a considerable amount of
storage room on the gadget.
12

With regards to landmark recognition applications, for example, storing information relating
to a large number of landmarks on a cell phone could seem infeasible.
However, in order to overcome this drawback, the developers of Monulens derived
inspiration from “maps” applications. Many maps applications allow the user to download
data that pertains only to the regions that they would be interested in.
This data is downloaded as an array of latitude, longitude values, topologies and contours.
Thus, in the absence of an active connection, the user can still localize their location on
the map, since GPS does not need a network connection.
While there is a certain image processing pipeline model that may serve as a guide and
starting point, it does not contain all of the answers for a situation where processing often
has to be split between local computing and remote, server operations (Girod et al., 2011)
Since the app created will be both a client-server application, three processing schema
need to be considered. In either case, there are certain performance limitations that the
application needs to avoid exceeding, lest it prove too resource-intensive to be a feasible
addition to a mobile suite of existing image recognition apps:

a) The mobile device sends an image to a server, where image-based retrieval is done
entirely via the remote server, including the analysis/matching of the image sent by
the mobile device.
b) The mobile device analyses and encodes the query image (local image features or
descriptors) on the device, after which it ends the extracted features to the server,
where only retrieval algorithms are executed.
c) The mobile device relies on a cache of data and carries out feature recognition all
on its own locally. The server is queried with a search request only if the device fails
to find a match (Girod et al., 2011).
14
Fig. 3: Three mobile visual search architectures (Girod et al., p.87)

Interestingly, in all three of the above cases the same principle applies; that of BoF or,
alternatively, BoW.
BoF stands for the bag-of-features approach (Yang et al., 2012), while BoW stands for
bag-of-words (Sivic and Zisserman, 2009; Chen and Yap, 2013; Farhangi, Soryani and
Fathy, 2014).
BoW is essentially an extension of text documentation retrieval, where a particular piece of
text (such as a webpage address) may be located through a few well-selected words typed
into a search engine (Girod et al., 2011; Sivic and Zisserman, 2009).
BoW may also be used to represent the document itself in a database, no matter where
particular words appear within the document. The only difference is that, for image
matching, local image features (those that stand out the most) are used in place of words
to perform the same matching function.
This should be possible to apply to landmark recognition in a relatively straightforward
fashion (Girod et al., 2011).
With BoF, it likewise does not matter where the features occur in the image, particularly in
the initial stages of landmark feature retrieval (Zitová and Flusser, 1999; Nanni and Lumini,
2013).
Still, features vary, which means that several different images of the same object may
need to be compared (Zhou, Zhou and Hu, 2013).
Thus, in conclusion, the decision as to whether this should be done locally or via a
remotely-accessed server may be determined by whether or not a landmark recognition
and mapping application may truly run properly on a mobile device (Chen, Yap, and
Zhang, 2014).
The present Android app development project aims to shed more light on this topic.

Author(s) Year Title Description Outcome
Colios, C.I. and
Trahanias, P.E. 2001
A framework
for visual
landmark
identification
based on
projective and
point-
permutation
invariant
vectors
Visual
landmark
identification
for robots and
autonomous
systems
Technique proved to
be useful in robot
navigation, therefore it
should be adaptable to
mobile use
Gillner, S., Weiß,
A.M. and Mallot,
H.A.
2008
Visual homing
in the absence
of feature-
based
landmark
information
Techniques for
image
recognition
based on
templates
rather than
features
Software exists for
recognising landmarks
and other elements
even without the use of
feature comparison
between multiple
images
Girod, B.,
Chandrasekhar, V.,
Grzeszczuk, R.
and Reznik, Y.A.
2011
Mobile visual
search:
Architectures,
technologies,
and the
emerging
MPEG
standard
Visual feature
recognition for
both images
and videos
tested on
mobile devices
There are multiple
ways of splitting
processing tasks
between the mobile
device itself and a
server, so long as
internet access is
available
Chan, E.C.L. and
Baciu, G. 2012
Introduction to
wireless
localisation:
With iPhone
SDK
examples
Wireless
localisation
techniques
based on
mobile phones
Feature-based
representation may be
done locally in only a
very limited fashion
compared to the cloud,
but the technique is
certainly viable
Aditya Srinivas
Timmaraju
&
Anirban Chatterjee 2013
Monulens:
real-time
mobile-based
landmark
recognition
Uses OpenCV
functions
utilising the
Java Native
Interface and
the Oriented
and Rotated
BRIEF
feature
descriptor
Cross device
validation returned
90% correct image
matching
Fig.4: Comparison of some of the most relevant sources
16

3.0 Summary
From reviewing a range of different studies about image recognition and carrying out
extensive research about different image recognition techniques and recognition devices
that are currently available (such as Monulens), some limitations relating to landmark
identification have been identified.
These include the following:
- Key point identification/descripiction and algorithms such as RANSAC (for evaluating
homography) appear to be computationally costly implementations.
- The accuracy of key points/descriptors and the real time computation of these is reduced
in the earlier approach.
- Comparisons against referenced images takes longer and is less efficient when there is
no mechanism for cache-blocking
The limitations identified above present the opportunity for further research and
development when creating an image recognition app through the implementation of:
- Algorithm Optimisation to overcome the problem in the key point detection outlined above
- The pre-computation of key points/descriptors of reference images so that it does not
have to repeat this step (thus reducing comparison time).
-The implementation of cache blocking and prefetching techniques which may have the
potential to reduce identification time and to improve overall performance.
Developing the proposed real-time landmark recognition app will also provide the
opportunity to discover what is involved in creating an efficient server-support based
application.

4.0 Conclusion
A substantial amount of research is available on visual pattern recognition in general, and
landmark recognition in particular.
However, there are certain issues in mobile application research that have not been
sufficiently studied and resolved, most important of which is when and how to split image
retrieval and processing tasks between the local mobile device and a remote server.
The real-time mobile Android app proposed by this project could become one of the few
optimised and widely-usable, location-aware applications that do not require a network to
function, but which enable such benefits as tagging photos and research data in Google
Maps, geo-tagging, and facilitating new experimental studies and field research where
geo-tagged resources are particularly important (Chan and Baciu, 2012).
18

5.0 References
Almasi, G.S. and Gottlieb, A. (1987). Highly parallel processing. Redwood City,
CA: Benjamin-Cummings Publishing Co., Subs. of Addison Wesley Longman, US.
Ansari, N. and Li, K. (1993). ‘Landmark-based shape recognition by a modified Hopfield
neural network’, Pattern Recognition, 26(4), pp. 531–542. doi: 10.1016/0031-
3203(93)90108-9.
Chen, D.M. (2013) Memory-Efficient Image Databases for Mobile Visual Search
Available at: https://www.youtube.com/watch?v=LlHaaVTB6S8
(Accessed: 27 November 2016).
Chan, E.C.L. and Baciu, G. (2012). Introduction to wireless localization. Singapore: IEEE
Chen, T. and Yap, K.-H. (2013). ‘Context-aware Discriminative vocabulary learning for
mobile landmark recognition’, IEEE Transactions on Circuits and Systems for Video
Technology, 23(9), pp. 1611–1621. doi: 10.1109/tcsvt.2013.2254978.
Chen, T., Yap, K.-H. and Zhang, D. (2014). ‘Discriminative soft bag-of-visual phrase for
mobile landmark recognition’, IEEE Transactions on Multimedia, 16(3), pp. 612–622.
doi: 10.1109/tmm.2014.2301978.
Colios, C.I. and Trahanias, P.E. (2001). ‘A framework for visual landmark identification
based on projective and point-permutation invariant vectors’, Robotics and Autonomous
Systems, 35(1), pp. 37–51. doi: 10.1016/s0921-8890(00)00129-9.
Elmogy, M. and Zhang, J. (2009). Landmark Recognition for Humanoid Robot
Navigation. International Conference on Robotics and Biomimetics Bangkok, Thailand,
Proceedings of the 2008 IEEE. [online] Available at:
http://ieeexplore.ieee.org/document/4913065/ [Accessed 3 Dec. 2016].
Farhangi, M.M., Soryani, M. and Fathy, M. (2014). ‘Informative visual words construction
to improve bag of words image representation', IET Image Processing, 8(5), pp. 310–318.
doi: 10.1049/iet-ipr.2013.0449.

Gillner, S., Weiß, A.M. and Mallot, H.A. (2008). ‘Visual homing in the absence of feature-
based landmark information’, Cognition, 109(1), pp. 105–122. doi:
10.1016/j.cognition.2008.07.018.
Girod, B., Chandrasekhar, V., Grzeszczuk, R. and Reznik, Y.A. (2011). ‘Mobile visual
search: Architectures, technologies, and the emerging MPEG standard’, IEEE Multimedia,
18(3), pp. 86–94. doi: 10.1109/mmul.2011.48.
Grama, A., Gupta, A. and Karypis, G. (2003). An introduction to parallel computing. 2nd
edn. Harlow, UK: Addison-Wesley Educational Publishers.
Hwang, K., Fox, G.C. and Dongarra, J.J. (2011). Distributed and cloud computing: From
parallel processing to the Internet of things. Amsterdam: Morgan Kaufmann Publishers Inc.
Landfeldt, B. (2009). ‘Special issue of the journal of parallel and distributed computing:
Advancement of research in wireless access and mobile systems’, Journal of Parallel and
Distributed Computing, 69(6), p. 588. doi: 10.1016/j.jpdc.2009.03.001.
Li, H. and Yang, S.X. (2003). ‘A behavior-based mobile robot with a visual landmark-
recognition system’, IEEE/ASME Transactions on Mechatronics, 8(3), pp. 390–400. doi:
10.1109/tmech.2003.816818.
Li, X., Wang, J., Knight, N. and Ding, W. (2011). ‘Vision-based positioning with a single
camera and 3D maps: Accuracy and reliability analysis’, Journal of Global Positioning
Systems, 10(1), pp. 19–29. doi: 10.5081/jgps.10.1.19.
Li, Z. and Yap, K.H. (2012). ‘Content and context boosting for mobile landmark
recognition’, IEEE Signal Processing Letters, 19(8), pp. 459–462.
doi: 10.1109/lsp.2012.2203120.
Mata, M., Armingol, J.M., Escalera, A. de la and Salichs, M.A. (2002). ‘Learning visual
landmarks for mobile robot navigation’, IFAC Proceedings Volumes, 35(1), pp. 445–450.
doi: 10.3182/20020721-6-es-1901.01299.
20

Meyer-Delius, D., Beinhofer, M., Kleiner, A. and Burgard, W. (2011).
Using artificial landmarks to reduce the ambiguity in the environment of a mobile robot.
IEEE International Conference on Robotics and Automation (ICRA), [online] pp.5173-
5178. Available at: https://www.researchgate.net/publication/221076556_Using
_artificial_landmarks_to_reduce_the_ambiguity_in_the_environment_of_a_mobile_robot
[Accessed 27 Nov. 2016].
Nanni, L. and Lumini, A. (2013). ‘Heterogeneous bag-of-features for object/scene
recognition’, Applied Soft Computing, 13(4), pp. 2171–2178. doi:
10.1016/j.asoc.2012.12.013.
Parker, J.R. (1996). CD-ROM to accompany Algorithms for image processing and
computer vision. New York: Wiley Computer Pub.
Pinto, N., Cox, D. and DiCarlo, J. (2008). Why is Real-World Visual Object Recognition
Hard?. PLOS COMPUTATIONAL BIOLOGY, 4(1), p.0151.
Rekhansh, R., Verma, T., Jaiswal, D., Professor, A. and Tech, M. (2015). ‘A survey on
landmark recognition techniques’, International Journal for Scientific Research &
Development|, 3, pp. 2321–613.
Rohr, K. (2010). Landmark-Based Image Analysis : Using Geometric and Intensity
Models. 1st ed. New York: Springer Publishing, pp.4.
Pablo Sala, Robert Sim, Ali Shokoufandeh and Sven Dickinson. “Landmark Selection
for Vision-Based Navigation” IEEE TRANSACTIONS ON ROBOTICS, VOL. 22, NO. 2,
APRIL 2006.
Sala, P., Sim, R., Shokoufandeh, A. and Dickinson, S. (2006). Landmark Selection for
Vision-Based Navigation. IEEE TRANSACTIONS ON ROBOTICS, 22(2).
Seo, J.S. and Yoo, C.D. (2004). ‘Localized image watermarking based on feature points of
scale-space representation’, Pattern Recognition, 37(7), pp. 1365–1375.
doi: 10.1016/j.patcog.2003.12.013.
Sivic, J. and Zisserman, A. (2009). ‘Efficient visual search of videos cast as text retrieval’,
IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(4), pp. 591–606.
doi: 10.1109/tpami.2008.111.

Yairi, T., Hirama, K. and Hori, K. (2003). ‘Map construction based on Cooccurrence
frequency of landmark observation’, Journal of the Robotics Society of Japan, 21(2), pp.
212–223. doi: 10.7210/jrsj.21.212.
Yang, F., Lu, H., Zhang, W. and Yang, G. (2012). ‘Visual tracking via bag of features’, IET
Image Processing, 6(2), p. 115. doi: 10.1049/iet-ipr.2010.0127.
Yan-Tao, Z., Ming, Z., Yang, S., Hartwig, A., Ulrich, B., Alessandro, B., Fernando, B.,
Tat-Seng, C. and Hartmut, N. (2009). Tour the World: building a web-scale landmark
recognition engine. 1st ed. [ebook]
Available at: http://www.google.com/googleblogs/pdfs/google_landmark_recognition.pdf
[Accessed 27 Nov. 2016].
Zhou, L., Zhou, Z. and Hu, D. (2013). ‘Scene classification using a multi-resolution bag-
of-features model’, Pattern Recognition, 46(1), pp. 424–433. doi:
10.1016/j.patcog.2012.07.017.
Zitová, B. and Flusser, J. (1999). ‘Landmark recognition using invariant features’, Pattern
Recognition Letters, 20(5), pp. 541–547. doi: 10.1016/s0167-8655(99)00031-8.
22

6.0 Bibliography
Rohr, K. (2010) Landmark-Based Image Analysis: Using Geometric and Intensity Models
(Computational Imaging and Vision). New York: Springer Publishing.
Treiber, M. (2010) An Introduction to Object Recognition: Selected Algorithms for a Wide
Variety of Applications (Advances in Computer Vision and Pattern Recognition.
New York: Springer Publishing.