3G MOBILE COMMUNICATION

&&&&&3GMOBILE COMMUNICATION&&&&&

3G network --> ABSTRACT

3rd Generation Wireless, or 3G, is the generic term used for the next generation of mobile communications systems. 3G systems aim to provide enhanced voice, text and data services to user. The main benefit of the 3G technologies will be substantially enhanced capacity, quality and data rates than are currently available.


3G Mobile will enable the provision of advanced services transparently to the end user and will bridge the gap between the wireless world and the computing/Internet world, making inter-operation apparently seamless.


The third generation networks should be in a position to support real-time video, high-speed multimedia and mobile Internet access. All this should be possible by means of highly evolved air interfaces, packet core networks, and increased availability of spectrum.


The ability to provide high-speed data is one of the key features of third generation networks, the real strength of these networks will be providing enhanced capacity for high quality voice services. The need for landline quality voice capacity is increasing more rapidly than the current 2nd generation networks will be able to support.


High data capacities will open new revenue sources for the operators and bring the Internet more closer to the mobile customer. The use of all-ATM or all-IP based communications between the network elements will also bring down the operational costs of handling both voice and data, in addition to adding flexibility.


The drive for 3G is the need for higher capacities and higher data rates. Whereas higher capacities can basically be obtained by having a greater chunk of spectrum or by using new evolved air interfaces, the data requirements can be served to a certain extent by overlaying 2.5G technologies on the existing networks. In many cases it is possible to provide higher speed packet data by adding few network elements and a software


The 3rd Generation Mobile System will most likely grow out of the convergence of enhanced 2nd generation mobile systems with greater data transfer speed and capacity and 1st generation satellite mobile systems. Evolution to the current generation mobile networks to 3G doesn't necessarily mean seamless up gradation to the existing infrastructure to the 3G.

3G --> CONTENTS

1. INTRODUCTION

2. A GLOBAL VISION TO 3G MOBILE

3. 3G ARCHITECTUE

4. CHARECTRISTICS OF 3G

5. EVOLUTION TOWARDS 3G

6. SECURITY ASPECTS

7. 3G SECURITY FEATURES

8. CONCLUSION

1. 3G --> Introduction

3rd Generation Wireless, or 3G, is the generic term used for the next generation of mobile communications systems. 3G systems aim to provide enhanced voice, text and data services to user. The main benefit of the 3G technologies will be substantially enhanced capacity, quality and data rates than are currently available.

3G Mobile will enable the provision of advanced services transparently to the end user and will bridge the gap between the wireless world and the Computing/Internet world, making inter-operation apparently seamless.

The third generation networks should be in a position to support real-time video, high-speed multimedia and mobile Internet access. All this should be possible by means of highly evolved air interfaces, packet core networks, and increased availability of spectrum.

2. 3G --> A Global Vision to 3G Mobile

“The IMT-2000 third generation mobile standard enables mobile users to harness the
full power of the Internet through efficient high-speed radio transmission, optimized for multimedia communications”

3. 3G Architecture

The 3G network will have a layered architecture, which will enable the efficient delivery of voice and data services. A layered network architecture, coupled with standardized open interfaces, will make it possible for the network operators to introduce and roll out new services quickly



These networks will have a connectivity layer at the bottom providing support for high quality voice and data delivery. Using IP or ATM or a combination of both, this layer will handle all data and voice info.

The layer consists of the core network equipment like routers, ATM switches and transmission equipment. Other equipment provides support for the core bit stream of voice or data, providing QOS etc.

In 3G networks, voice and data will not be treated separately which could lead to a reduction in operational costs of handling data separately from voice. The application layer on top will provide open application service interfaces enabling flexible service creation. This user application layer will contain services for which the end user will be willing to pay. These services will include eCommerce, GPS and other differentiating services. In between the application layer and the connectivity layer, will run the control layer with MSC servers, support servers, HLR etc. These servers are needed to provide any service to a subscriber.

4. 3G --> Characteristics of 3G

The main characteristics of 3G systems, known collectively as IMT–2000, are a single family of compatible standards that have the following characteristics:

• Used worldwide

• Used for all mobile applications

• Support both packet-switched (PS) and circuit-switched (CS) data transmission

• Offer high data rates up to 2 Mbps (depending on mobility/velocity)

• Offer high spectrum efficiency

Figure 1. Multiple Standards for Different Applications and Countries



IMT–2000 is a set of requirements defined by the International Telecommunications Union (ITU). As previously mentioned, IMT stands for International Mobile Telecommunications, and “2000” represents both the scheduled year for initial trial systems and the frequency range of 2000 MHz. All 3G standards have been developed by regional Standards Developing Organizations (SDOs). In total, proposals for 17 different IMT–2000 standards were submitted by regional SDOs to ITU in 1998—11 proposals for terrestrial systems and 6 for Mobile Satellite Systems (MSSs). Evaluation of the proposals was completed at the end of 1998, and negotiations to build a consensus among differing views were completed in mid 1999. All 17 proposals have been accepted by ITU as IMT–2000 standards. The specification for the Radio Transmission Technology (RTT) was released at the end of 1999.

The most important IMT–2000 proposals are the UMTS (W-CDMA) as the successor to GSM, CDMA2000 as the interim standard ’95 (IS–95) successor, and time division–synchronous CDMA (TD–SCDMA) (universal wireless communication–136 [UWC–136]/EDGE) as TDMA–based enhancements to D–AMPS/GSM—all of which are leading previous standards toward the ultimate goal of IMT–2000.

UMTS allows many more applications to be introduced to a worldwide base of users and provides a vital link between today’s multiple GSM systems and IMT–2000. The new network also addresses the growing demand of mobile and Internet applications for new capacity in the overcrowded mobile communications sky. UMTS increases transmission speed to 2 Mbps per mobile user and establishes a global roaming standard.

UMTS is being developed by Third-Generation Partnership Project (3GPP), a joint venture of several SDOs—ETSI (Europe), Association of Radio Industries and Business/Telecommunication Technology Committee (ARIB/TTC) (Japan), American National Standards Institute (ANSI) T-1 (USA), telecommunications technology association (TTA) (South Korea), and Chinese Wireless Telecommunication Standard (CWTS) (China). To reach global acceptance, 3GPP is introducing UMTS in phases and annual releases. The first release (UMTS Rel. ’99), introduced in December of 1999, defines enhancements and transitions for existing GSM networks. For the second phase (UMTS Rel. ’00), similar transitions are being proposed as enhancements for IS–95 (with CDMA2000) and TDMA (with TD–CDMA and EDGE).

The most significant change is the new UMTS Terrestrial Radio Access (UTRA), a W–CDMA radio interface for land-based communications. UTRA supports time division duplex (TDD) and frequency division duplex (FDD). The TDD mode is optimized for public micro and pico cells and unlicensed cordless applications. The FDD mode is optimized for wide-area coverage, i.e., public macro and micro cells. Both modes offer flexible and dynamic data rates up to 2 Mbps. Another newly defined UTRA mode, multicarrier (MC), is expected to establish compatibility between UMTS and CDMA2000.

5. 3G --> Evolution towards 3G

Phase 1 of the standardization of GSM900 was completed by the European Telecommunications Standards Institute (ETSI) in 1990 and included all necessary definitions for the GSM network operations. Several tele services and bearer services have been defined (including data transmission up to 9.6 kbps), but only some very basic supplementary services were offered. As a result, GSM standards were enhanced in Phase 2 (1995) to incorporate a large variety of supplementary services that were comparable to digital fixed network integrated services digital network (ISDN) standards. In 1996, ETSI decided to further enhance GSM in annual Phase 2+ releases that incorporate 3G capabilities.

GSM Phase 2+ releases have introduced important 3G features such as intelligent network (IN) services with customized application for mobile enhanced logic (CAMEL), enhanced speech compression/decompression (CODEC), enhanced full rate (EFR), and adaptive multirate (AMR), high–data rate services and new transmission principles with high-speed circuit-switched data (HSCSD), general packet radio service (GPRS), and enhanced data rates for GSM evolution (EDGE). UMTS is a 3G GSM successor standard that is downward-compatible with GSM, using the GSM Phase 2+ enhanced core.


The swift migration from 2G to 3G and their bandwidth used are as in the Figure.

6. 3g --> Security Aspects

• Network Authentication
The user can identify the network
• Explicit Integrity
Data integrity is assured explicitly by use of integrity algorithms
Also stronger confidentiality algorithms with longer keys
• Network Security
Mechanisms to support security within and between networks


• Switch Based Security
Security is based within the switch rather than the base station
• IMEI Integrity
Integrity mechanisms for IMEI provided from the start
• Secure Services
Protect against misuse of services provided by SN and HE
• Secure Applications
Provide security for applications resident on USIM
• Fraud Detection
Mechanisms to combating fraud in roaming situations
• Flexibility
Security features can be extended and enhanced as required by new threats and services
• Visibility and Configurability
Users are notified whether security is on and what level of security is available.
Users can configure security features for individual services

7. 3G Security Features

Mutual Authentication
During Authentication and Key Agreement (AKA) the user and network authenticate each other, and also they agree on cipher and integrity key (CK, IK). CK and IK are used until their time expires.
Assumption: trusted HE and SN, and trusted links between them.
After AKA, security mode must be negotiated to agree on encryption and integrity algorithm.

8. 3G Network --> Conclusion

Standardization of 3G mobile systems is based on ITU (International Telecom Union) recommendations for IMT 2000. IMT 2000 specifies a set of requirements that must be achieved 100% for a network to be called 3G. By providing multimedia capacities and higher data rates, these systems will enhance the range and quality of services provided by 2G systems.
The main contenders for 3G systems are wideband CDMA (W-CDMA) and CDMA2000. The ETSI/ GSM players including infrastructure vendors such as Nokia and Ericsson backed W-CDMA. cdma2000 was backed by the North American CDMA community, led by the CDMA Development Group (CDG) including infrastructure vendors such as Qualcomm and Lucent Technologies. Universal Mobile Telephone System (UMTS) is the widely used European name for 3G.
The proposed IMT-2000 standard for third generation mobile networks globally is a CDMA-based standard that encompasses THREE OPTIONAL modes of operation, each of which should be able to work over both GSM MAP and IS-41 network architectures.
UMTS is the European designation for 3G systems. The UMTS frequency bands selected by the ITU are 1,885 MHz - 2,025 MHz (Tx) and 2,110 MHz - 2,2,20 MHz (Rx). Higher frequency bands could be added in future if need be, for stationary data. There is still some confusion about all the frequency options as FCC has not given clear indications so far. The following table should briefly give an idea about the 3G system specifications.

3D-FACE RECOGNITION

3D-Face Recognition Using The Fishersurface Method And Surface Space Combinations

Abstract

In this paper we test a range of three-dimensional face recognition systems, based on the fishersurface method developed in previous work. We show the effect of using a variety of facial surface representations and suggest a method of identifying and extracting useful qualities offered by each system. Combing these components into a unified surface subspace, we create a three dimensionalface recognition system producing significantly lower error rates than individual systems tested on the same data. We evaluate systems by performing up to 1,079,715 verification operations on a large test set of 3D face models. Results are presented in the form of false acceptance and false rejection rates, generated by varying a decision threshold applied to a distance metric in combined surface space.

3D Faces --> INGREDIENTS

o Introduction

o Fisher surface method

o Test database

o Surface space analysis

o Combining systems-multiple face recognition

o Test procedure

o Results

o Conclusion

3D Faces --> Introduction

Despite significant advances in face recognition technology, it has yet to achieve levels of accuracy required for many commercial and industrial applications. The high error rates stem from well-known sub-problems. Variation in lighting, facial expression and orientation all significantly increase error rates. In an attempt to address these issues, research has begun to focus on the use of three-dimensional face models, motivated by three main factors. Firstly, relying on geometric shape, rather than colour and texture information, systems become invariant to lighting conditions. Secondly, the ability to rotate a facial structure in three-dimensional space, allowing for compensation of variations in pose, aids those methods requiring alignment prior to recognition. Thirdly, the additional depth information in the facial surface structure, not available from twodimensional images, provides supplementary cues for recognition.

In this paper we expand on previous research involving the use of facial surface data, derived from 3D face models (generated using a stereo vision 3D camera), as a substitute for the more familiar two-dimensional images. A number of investigations have shown that three-dimensional structure can be used to aid recognition. Zhao and Chellappa use a generic 3D face model to normalise facial orientation and lighting direction in two-dimensional images, increasing recognition accuracy from
approximately 81% (correct match within rank of 25) to 100%. Similar results are witnessed in the Face Recognition Vendor Test , showing that pose correction using Romdhani et al’s technique reduces error rates when applied to the FERET database.

Blanz et al [5] take a comparable approach, using a morphable face model to aid in identification of 2D images. Beginning with an initial estimate of lighting direction and face shape, Romdhani et al iteratively alters shape and texture parameters of the morphable face model, minimising difference to the two-dimensional image. These parameters are then taken as features for identification, resulting in 82.6% correct identifications on a test set of 68 people.
Although these methods show that knowledge of three-dimensional face shape can aid normalisation for two-dimensional face recognition systems, none of the methods mentioned so far use actual three-dimensional geometric structure to perform recognition. Whereas Beumier and Acheroy [6, 7] make direct use of such information, testing various methods of matching 3D face models, although few were successful.

Curvature analysis proved ineffective, and feature extraction was not robust enough to provide accurate recognition. However, Beumier and Acheroy were able to achieve reasonable error rates using curvature values of vertical surface profiles. Verification tests carried out on a database of 30 people produced equal error rates (EER) between 7.25% and 9.0%. Hesher et al test a different method, using PCA (principal component analysis) of depth maps and euclidean distance to perform identification with 94% accuracy on 37 face models (when trained on the gallery set). Further investigation into this approach is carried out by Heseltine et al, showing how different surface representations and distance measures affect recognition, reducing the EER from 19.1% to 12.7% when applied to a difficult test set of 290 face models. However, the focus of this research has been on identifying optimum surface representations, with little regard for the advantages offered by each individual representation.

We suggest that different surface representations may be specifically suited to different capture conditions or certain facial characteristics, despite a general weakness for overall recognition. For example, curvature representations may aid recognition by making the system more robust to inaccuracies in 3D orientation yet also be highly sensitive to noise. Another representation may enhance nose shape, but lose information regarding jaw structure.

In this paper we analyse and evaluate a variety of three-dimensional fishersurface face recognition systems, each incorporating a different surface representation of facial structure. We propose a means of identifying and extracting components from the surface subspace produced by each system, such that they may be combined into a single unified subspace. Pentland et al [10] have previously examined the benefit of using multiple eigenspaces, in which specialist subspaces were constructed for various facial orientations, from which cumulative match scores were able to reduce error rates. Our approach differs in that we extract and combine individual dimensions, creating a single unified surface space, as applied to two-dimensional images in previous investigations.

3D Faces --> The Fishersurface Method

In this section we provide details of the fishersurface method of face recognition. We apply PCA and LDA (linear discriminant analysis) to surface representations of 3D face models, producing a subspace projection matrix, as with Belhumier et al’s fisherface approach [12], taking advantage of ‘within-class’ information, minimising variation between multiple face models of the same person, yet maximising class separation. To accomplish this we use a training set containing several examples of each subject, describing facial structure variance (due to influences such as facial expression), from one model to another. From the training set we compute three scatter matrices, representing the within-class (SW), between-class (SB) and total (ST) distribution from the average surface _ and classes averages _n, as shown in equation 1.

The training set is partitioned into c classes, such that all surface vectors ni in a single class Xn are of the same person and no person is present in multiple classes. Calculating eigenvectors of the matrix ST, and taking the top 250 (number of surfaces minus number of classes) principal components, we produce a projection matrix Upca. This is then used to reduce dimensionality of the within-class and between-class scatter matrices (ensuring they are non-singular) before computing the top c-1 eigenvectors of the reduced scatter matrix ratio, Ufld, as shown in equation 2.

Finally, the matrix Uff is calculated, such that it projects a face surface vector into a reduced space of c-1 dimensions, in which between-class scatter is maximised for all c classes, while within-class scatter is minimised for each class Xn. Like the fisherface system [12], components of the projection matrix Uff can be viewed as images, as shown in Figure 1 for the depth map surface space.



Figure 1: The average surface (left) and first five fishersurfaces (right)
Once surface space has been defined, we project a facial surface into reduced surface
space by a simple matrix multiplication, as shown in equation 3.


^=(p−!)TU (3)

The vector _T___1__2____c-1] is taken as a ‘face-key’ representing the facial structure in reduced dimensionality space. Face-keys are compared using either euclidean or cosine distance measures as shown in equation 4.



An acceptance (facial surfaces match) or rejection (surfaces do not match) is determined by applying a threshold to the distance calculated. Any comparison producing a distance value below the threshold is considered an acceptance.

3D Face --> The Test Database

Until recently, little three-dimensional face data has been publicly available for research and nothing towards the magnitude required for development and testing of threedimensional face recognition systems. In these investigations we use a new database of 3D face models, recently made available by the University of York, as part of an ongoing project to provide a publicly available 3D Face Database . Face models are generated in sub-second processing time from a single shot with a 3D camera, using a stereo vision technique enhanced by light projection.

For the purpose of these experiments we select a sample of 1770 face models (280 people) captured under the conditions in Figure 2. During data acquisition no effort was made to control lighting conditions. In order to generate face models at various head orientations, subjects were asked to face reference points positioned roughlyabove and below the camera, but no effort was made to enforce precise orientation.


3D models are aligned to face directly forwards before conversion into 60 by 90 pixel depth map representation. We then take a training set of 300 depth maps (50 people), used to compute the scatter matrices described in section 3. The remaining 1470 depth maps (230 people) are then separated into two disjoint sets of equal size (test set A and test set B). We use test set A to analyse the face-key variance throughout surface space, calculate discriminant weightings (see section 4) and compute the
optimum surface space combinations. This leaves set B as an unseen test set to evaluate the final combined system. Both training and test sets contain subjects of various race, age and gender and nobody is present in both the training and test sets.



Figure 2: Example face models taken from the University of York 3D Face Database

3D Face --> Surface Space Analysis

In this section we analyse the surface spaces produced when various facial surface representations are used with the fishersurface method. We begin by testing the variety of fishersurface systems introduced by Heseltine et al [1] on test set A, showing the range of error rates produced when using various surface representations (Figure 3). Continuing this line of research we persist with the same surface representations, referring the reader to previous work [1, 9] for implementation details, while in this
paper we focus on the effect and methodologies of combining multiple systems, rather than the surface representations themselves. Figure 3 clearly shows the choice of surface representation has a significant impact on the effectiveness of the fishersurface approach, with horizontal gradient
representations providing the lowest EER (point at which false acceptance rate equals
false rejection rate).

However, the superiority of the horizontal gradient representations does not suggest that the vertical gradient and curvature representations are no use whatsoever. Although discriminatory information provided by these representations may not be as robust and distinguishing, they may contain a degree of information not available in horizontal gradients and could therefore still make a positive contribution to a combined surface space. We measure the discriminating ability of surface space dimensions by applying Fisher’s Linear Discriminant (FLD) (as used by Gordon [14]) to individual components
(single dimensions) of each surface space. We calculate the discriminant dn, describing
the discriminating power of a given dimension n, between c people in test set A.



Where _i is the set of all class i face-key vector elements in dimension n, and m and mi the mean and class mean of nth dimension elements in test set A. Applying equation 3 to the assortment of surface space systems listed in Figure 3, we see a wide range of discriminant values across the individual surface space dimensions (Figure 4).




Figure 4: Top ten discriminant values of all fishersurface dimensions

It is clear that although some surface representations do not perform well in the face recognition tests, producing high EERs, some face-key components do contain highly discriminatory information. For example, we see that the min and max curvature representations contain one dimension with a higher discriminant than any horizontal gradient and curve type dimension, yet the EERs are significantly higher. We hypothesise that the reason for these highly discriminating anomalies, in an otherwise ineffective subspace, is that a certain surface representation may be particularly suited to a single discriminating factor, such as nose shape or jaw structure, but is not effective when used as a more general classifier. Therefore, if we were able to isolate these few useful qualities from the more specialised subspaces, they could be used to make a positive contribution to a generally more effective surface space, reducing error rates further.

3D Face --> Combining Systems

In this section we describe how the analysis methods discussed in section 4 are used to combine multiple face recognition systems. Firstly, we need to address the problem of prioritising surface space dimensions. Because the average magnitude and deviation of face-key vectors from a range of systems are likely to differ by some orders of magnitude, certain dimensions will have a greater influence than others will, even if the discriminating abilities are evenly matched. To compensate for this effect, we normalize moments by dividing each face-key element by its within-class standard deviation (calculated from test set A face-keys). However, in normalising these dimensions we have also removed any prioritisation, such that all surface space components are considered equal. Although not a problem when applied to a single surface space, when combining multiple dimensions we would ideally wish to give greater precedence to the more reliable components. Otherwise the situation is likely to arise when a large number of less discriminating dimensions begin to outweigh the fewer more discriminating dimensions, diminishing their influence on the verification operation and hence increasing error rates.

In section 4 we showed how FLD could be used to measure the discriminating ability of a single dimension from any given face space. We now apply this discriminant value dn as weighting for each surface space dimension n, prioritising those dimensions with the highest discriminating ability.
With this weighting scheme applied to all face-keys, we now require some criterion to decide which dimensions to combine. It is not enough to rely purely on the discriminant value itself, as this only provides an indication of the discriminating ability of that dimension alone, without any indication of whether the inclusion of this dimension would benefit the existing set of dimensions. If an existing surface space already provides a certain amount of feature specific discrimination, it would be of little
benefit (or could even be detrimental) if we were to introduce an additional dimension describing a feature already present within the existing set.

Previous investigations have used FLD, applied to a combined subspace in order to predict effectiveness when used for recognition. Additional dimensions are introduced if they result in an increase in discriminant value. This method has been shown to produce face space combinations achieving significantly lower error rates than individual two-dimensional systems, although Heseltine et al do note that an EER-based criterion is likely to produce a better combination, at the expense of greatly increased training time. However, with a more efficient program and greater computational resources, we now take that approach: the criterion required for introduction of a new dimension to an existing surface space is a resultant decrease in EER (computed using test set A).


Combined surface space = first dimension of current optimum system
Compute EER of combined surface space
For each surface representation system:
For each dimension of surface space:
Concatenate dimension onto combined surface space
Compute EER of combined surface space
If EER has not decreased:
Remove dimension from combined surface space
Save combined surface space ready for evaluation


Figure 5: Fishersurface combination algorithm

3D Face --> The Test Procedure

In order to evaluate the effectiveness of a surface space, we project and compare each facial surface with every other surface in the test set, no surface is compared with itself and each pair is compared only once. The false acceptance rate (FAR) and false rejection rate (FRR) are then calculated as the percentage of incorrect acceptances and incorrect rejections after applying a threshold. By varying the threshold, we produce a series of FAR FRR pairs, which plotted on a graph produce an error curve as seen in Figure 8. The equal error rate (EER, the point at which FAR equals FRR) can then be
taken as a single comparative value.

3D Face--> Results

In this section we present the dimensions selected to form the combined fishersurface
systems (Figure 7) and the error rates obtained from a range of tests sets, making a
comparison to optimum individual systems in Figure 8.



We see that systems with lower EERs generally make the most contribution to the combined system, as would be expected. However, it is also interesting to note that even systems with particularly high EERs do contain some dimensions that make a positive contribution, although this is much more prominent for the cosine distance, showing that this metric is more suited to combing multiple surface spaces.
Having selected and combined the range of dimensions shown in Figure 7, we now apply these ombined systems to test sets A and B using both the cosine and Euclidean distance metric. We also perform an evaluation on the union of test sets A and B: an experiment analogous to training on a database (or gallery set) of known people, which are then compared with newly acquired (unseen) images.

Figure 8 shows the error curves obtained when optimum individual fishersurface systems and combined systems are applied to test set A (used to construct the combination), test set B (the unseen test set) and the full test set (all surfaces from sets A and B), using the cosine and Euclidean distance metrics. We see that the combined systems produce lower error rates than the optimum individual systems for all six
experiments. As would be expected, the lowest error rates are achieved when tested on the surfaces used to construct the combination (7.2% and 12.8% EER respectively).

However an improvement is also seen when applied to the unseen test set B, from 11.5% and 17.3% using the best single systems to 9.3% and 16.3% EER for the combined systems. Performing the evaluation on the larger set, providing 1,079,715 verification operations (completed in 14 minutes 23 seconds on a Pentium III 1.2GHz processor, providing a verification rate of 1251 per second), the error drops slightly to 8.2% and 14.4% EER, showing that a small improvement is introduced if some test data is available for training, as well as suggesting that the method scales well, considering the
large increase in verification operations.

Conclusion

We have shown how a well-known method of two-dimensional face recognition can be applied to three-dimensional face models achieving reasonably low error rates, depending on the surface representation used. Drawing on previous work combing face recognition eigenspaces [11], we have applied the same principle to multiple threedimensional face recognition systems, showing that the combination method is
applicable to both two-dimensional and three-dimensional data. Using FLD as an analysis tool, we have confirmed the hypothesis that although some surface representations may not perform well when used for recognition, they may harbor highly discriminatory components that could complement other surface spaces. Iteratively improving error rates on a small test set, we have built up a combination of dimensions extracted from a variety of surface spaces, each utilising a different surface representation. This method of combination has been shown to be most effective when used with the cosine distance metric, in which a selection of 184 dimensions were combined from 16 of the 17 surface spaces, reducing the EER from 11.6% to 8.2%. Applying the same combined surface space to an unseen test set of data presenting typical difficulties when performing recognition, we have demonstrated a similar reduction in error from 11.5% to 9.3% EER.

Evaluating the combined system at its fundamental level, using 1,079,715 verification operations between three-dimensional facial surfaces, demonstrates that combining multiple surface space dimensions improves effectiveness of the core recognition algorithm. Error rates have been significantly reduced to state-of-the-art levels, when evaluated on a difficult test set including variations in expression and orientation. However, we have not applied any additional heuristics, typically incorporated into fully functional commercial and industrial systems. For example, we have not experimented with multiple facial alignments, optimising crop regions or storing multiple gallery images. All of which are known to improve error rates and can easily be applied to the combined systems presented in this paper. With these additional measures in place, it is likely that the improvements made to the core algorithm will
propagate through to producing a highly effective face recognition system. Given the fast 3D capture method, small face-keys of 184 vector elements (allowing extremely fast comparisons), invariance to lighting conditions and facial orientation, this system is particularly suited to security and surveillance applications.

3D Face: References

 www-users.cs.york.ac.uk/~nep/research/3Dface/tomh/ 3DFaceRecUsingSurfaceSpaceCombinations-BMVC.pdf

 citeseer.ist.psu.edu/716122.html

 ieeexplore.ieee.org/iel5/4378863/4378864/04378912.pdf?arnumber= 4378912

 Heseltine, T., Pears, N., Austin, J. Three-Dimensional Face Recognition: A
Fishersurface Approach.