Abstract. Essentially, computer vision is the computer'

Added on - 20 Sep 2019

  • 23

    pages

  • 5175

    words

  • 132

    views

  • 0

    downloads

Showing pages 1 to 6 of 23 pages
AbstractEssentially, computer vision is the computer's ability to gain a high-level understandingfrom 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 ofan image as accurately and efficiently as possible remains both a primary interest and anexacting task within the domain of computer vision.Keywords:-Image Recognition, Machine Learning, Data-set,Bag-of-Words,GPSReceiver, Key Point Detection/Description,PICASA, Monulens and Google Goggles.1
Table of Contents1.0 Introduction......................................................................................................................42.0 Literature Review:...........................................................................................................62.1 A Survey of Major Image/Landmark Recognition & Comparison Techniques:..............62.2 Mobile Visual Search Architectures:.............................................................................113.0 Summary........................................................................................................................184.0 Conclusion.....................................................................................................................195.0 References.....................................................................................................................206.0 Bibliography...................................................................................................................24
3
1.0 IntroductionA touristic landmark is an instantly recognisable building or site (such as a monument or acathedral).As a traveller, gathering real-time, interactive andrelevant information about a landmark or monument andits surrounding areas of interest are important aspects ofan individual's journey because of the cultural andhistorical aspects itcan present.The World Wide Web is readily abundant with imagesand video recordings.Creating a mobile application which has the ability torecognise and match a vast number of landmarksefficiently still remains a challenge due to the sheer number of images that are required tobe searched within a database/dataset, along with the presence of possible visualdistortions such as external litter, illuminated amendment and dynamic geometry of theimaging 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 Weband the advancement of landmark picture sharing via websites such as Picasa and Flickr,the requirement for a computer conception to recognise landmarks universally through thecreation of reliable image identification engines and algorithms is necessary (Rohr, 2010).In this research work, a literature review will be conducted to assess researchdevelopments in the field of image recognition and the discoveries and progress that hasbeen 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 somepotential research directions and techniques that can be further explored and improvedupon identified.4Fig.1: Famous Landmarks (Google.com,2016)
Only through this process may an application be bestowed with image recognitioncapabilities advanced enough to fully recognise and understand a diverse range of imagesusing neural networks (Parker, 1996).This includes taking into consideration a number of issues that often present themselveswhen creating such a device, such as the variations that are guaranteed to emerge in anygiven 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 recognitionAndroid app.The device will use a large dataset of images which will be stored on the device itself formatching. By identifying a given landmark, the application will simplify both contentunderstanding and the geolocation of images returned via GPS, enabling a topographicalrepresentation and navigation of landmarks in the local area to provide appropriate tourguide suggestions and guidance via online resources such as Yelp and Google Maps.However, as already highlighted, efficiency is a challenge for sucha potentially unstable,large-scale image recognition system (Pinto, Cox and DiCarlo, 2008).
2.0 Literature Review:2.1 ASurveyof Major Image/Landmark Recognition & Comparison Techniques:Various works have been carried out within the area of image recognition adhering to therequirements of the main specifications.On observing the dearth of methods for the identification and correlation of specific detailsfrom 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 oflandmark recognition and modelling (some on a global-scale).Colios etal. (2001) experimented with automated landmark identification for robots bymeans of both projective and point-permutation invariant vectors. These vectors wereused to identify landmark patterns based on workspace planar features, enabling thecreation of direct, point-to-point correspondences in an indoor setting.This, in turn, allowed for the use of bothprojectivity constraintsand theconvex hulltoidentify 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'seffectiveness to be restrictive (Rekhansh et al., 2015).As part of a different project, vision-based landmark recognition was also used by Mata etal.(2002) toperform topological localisation of mobile robots.Both natural and artificial landmarks alike were used here, enabling the creation of asearch function whereby pattern identification/recognition techniques were applied todigital images.6
desklib-logo
You’re reading a preview
card-image

To View Complete Document

Become a Desklib Library Member.
Subscribe to our plans

Download This Document