Decision Fusion

Two research strands, each identifying an area of markedly increasing importance in the current development of pattern analysis technology, underlie the review covered by this paper, and are drawn together to offer both a task-oriented and a fundamentally generic perspective on the discipline of pattern recognition. The first of these is the concept of decision fusion for high-performance pattern recognition, where (often very diverse) classification technologies, each providing complementary sources of information about class membership, can be integrated to provide more accurate, robust and reliable classification decisions. The second is the rapid expansion in technology for the automated analysis of (especially) handwritten data for OCR applications including document and form processing, pen-based computing, forensic analysis, biometrics and security, and many other areas, especially those which seek to provide online or offline processing of data which is available in a human-oriented medium. Classifier combination/multiple expert processing has a long history, but the sheer volume and diversity of possible strategies now available suggest that it is timely to consider a structured review of the field. Handwritten character processing provides an ideal context for such a review, both allowing engagement with a problem area which lends itself ideally to the performance enhancements offered by multi-classifier configurations, but also allowing a clearer focus to what otherwise, because of the unlimited application horizons, would be a task of unmanageable proportions. Hence, this paper explicitly reviews the field of multiple classifier decision combination strategies for character recognition, from some of its early roots to the present day. In order to give structure and a sense of direction to the review, a new taxonomy for categorising approaches is defined and explored, and this both imposes a discipline on the presentation of the material available and helps to clarify the mechanisms by which multi-classifier configurations deliver performance enhancements. The review incorporates a discussion both of processing structures themselves and a range of important related topics which are essential to maximise an understanding of the potential of such structures. Most importantly, the paper illustrates explicitly how the principles underlying the application of multi-classifier approaches to character recognition can easily generalise to a wide variety of different task domains.

This paper describes a novel hand gesture recognition system that utilizes both multi-channel surface electromyogram (EMG) sensors and 3D accelerometer (ACC) to realize user-friendly interaction between human and computers. Signal segments of meaningful gestures are determined from the continuous EMG signal inputs. Multi-stream Hidden Markov Models consisting of EMG and ACC streams are utilized as decision fusion method to recognize hand gestures. This paper also presents a virtual Rubik's Cube game that is controlled by the hand gestures and is used for evaluating the performance of our hand gesture recognition system. For a set of 18 kinds of gestures, each trained with 10 repetitions, the average recognition accuracy was about 91.7% in real application. The proposed method facilitates intelligent and natural control based on gesture interaction.

The purpose of this paper is to present a view of a telemonitoring system based on fuzzy logic for distress situations detection of elderly people living alone in the home environment. This system includes three different sensors: physiological sensors (cardiac frequency, activity or agitation, posture and fall detection sensor), microphones and infrared sensors. Taking into account the difficulty of the statistical modeling of abnormal situation and considering that the Fuzzy Logic has been actually used with success in the development of classifier and analysis of control systems, its use in the decision fusion module of our home elderly people monitoring system is presented and evaluated in this paper.

Data fusion and decision fusion classification strategies are introduced to predict dyslexia from multi-channel event related potentials (ERPs) recorded, at birth, in response to multiple stimuli. Two data and two decision fusion strategies are developed in conjunction with nearest-mean classification rank selection to classify multi-stimuli multi-channel (MSMC) ERPs. The fusion vector in the data fusion strategy is formed by directly combining the rank-ordered MSMC ERP vectors or the rankordered elements of the MSMC ERPS. The resulting fusion vector is classified using a vector nearest-mean classifier. The nearest-mean classification decisions of the rank-ordered MSMC ERP vectors or the rank-ordered MSMC ERP elements are combined into a fusion vector in the decision fusion strategy. The resulting decision fusion vector is classified using a discrete Bayes classifier. The MSMC fusion classification strategies are tested on the averaged ERPs recorded at birth of 48 children: 17 identified as dyslexic readers, 7 as poor readers, and 24 identified as normal readers at 8 years of age. The ERPs were recorded at 6 electrode sites in response to two speech sounds and two non-speech sounds. It is shown that through the MSMC ERP element decision fusion strategy, dyslexic readers and poor readers can be predicted with almost 100% accuracy. Consequently, future reading problems can be detected early using neonatal responses making it possible to introduce more effective interventions earlier to children with reading problems emerging later in their lives. Furthermore, it is noted that because of the generalized formulations, the fusion strategies introduced can be applied, in general, to problems involving the classification of multi-category multi-sensor signals.

These days, with the increasing demands and responsibilities of life, parents becoming busy most of the time which makes taking care of an infant incredibly difficult task as infants need the attention and supervision of caregivers all the time. However, the advancement of technology has led to the emergence of smart baby cradle as an intelligent system with numerous features to reduce the burden of parents and caregivers. The paper aims to enhance the quality of the existing baby cradle systems by incorporating a new module for managing baby cry. Managing baby cry involves providing analysis for baby cry, and accordingly, triggering the suitable device attached to the cradle. This research paper provides significant attention on detecting baby cry, more accurately, by integrating four-sub modules in the cry classification process including voice analysis, face image analysis, body gesture analysis, and finally decision fusion.

We propose a binary decision fusion rule that reaches a global decision on the presence of a target by integrating local decisions made by multiple sensors. Without requiring a priori probability of target presence, the fusion threshold bounds derived using Chebyshev's inequality ensure a higher hit rate and lower false alarm rate compared to the weighted averages of individual sensors. The Monte Carlo-based simulation results show that the proposed approach significantly improves target detection performance, and can also be used to guide the actual threshold selection in practical sensor network implementation under certain error rate constraints.

Real-time implementation of an assistive human-machine interface system based around tongue-movement ear pressure (TMEP) signals is presented, alongside results from a series of simulated control tasks. The implementation of this system into an online setting involves short-term energy calculation, detection, segmentation and subsequent signal classification, all of which had to be reformulated based on previous off-line testing. This has included the formulation of a new classification and feature extraction method. This scheme utilises the discrete cosine transform to extract the frequency features from the time domain information, a univariate Gaussian maximum likelihood classifier and a two phase cross-validation procedure for feature selection and extraction. The performance of this classifier is presented alongside a real-time implementation of the decision fusion classification algorithm, with each achieving 96.28% and 93.12% respectively. The system testing takes into consideration potential segmentation of false positive signals. A simulation mapping commands to a planar wheelchair demonstrates the capacity of the system for assistive robotic control. These are the first real-time results published for a tongue-based human-machine interface that does not require a transducer to be placed within the vicinity of the oral cavity.

The classification of multisensor data sets, consisting of multitemporal synthetic aperture radar data and optical imagery, is addressed. The concept is based on the decision fusion of different outputs. Each data source is treated separately and classified by a support vector machine (SVM). Instead of fusing the final classification outputs (i.e., land cover classes), the original outputs of each SVM discriminant function are used in the subsequent fusion process. This fusion is performed by another SVM, which is trained on the a priori outputs. In addition, two voting schemes are applied to create the final classification results. The results are compared with well-known parametric and nonparametric classifier methods, i.e., decision trees, the maximumlikelihood classifier, and classifier ensembles. The proposed SVM-based fusion approach outperforms all other approaches and significantly improves the results of a single SVM, which is trained on the whole multisensor data set.

We present a speech signal driven emotion recognition system. Our system is trained and tested with the INTERSPEECH 2009 Emotion Challenge corpus, which includes spontaneous and emotionally rich recordings. The challenge includes classifier and feature sub-challenges with five-class and two-class classification problems. We investigate prosody related, spectral and HMM-based features for the evaluation of emotion recognition with Gaussian mixture model (GMM) based classifiers. Spectral features consist of mel-scale cepstral coefficients (MFCC), line spectral frequency (LSF) features and their derivatives, whereas prosody-related features consist of mean normalized values of pitch, first derivative of pitch and intensity. Unsupervised training of HMM structures are employed to define prosody related temporal features for the emotion recognition problem. We also investigate data fusion of different features and decision fusion of different classifiers, which are not well studied for emotion recognition framework. Experimental results of automatic emotion recognition with the INTER-SPEECH 2009 Emotion Challenge corpus are presented.

We propose a binary decision fusion rule that reaches a global decision on the presence of a target by integrating local decisions made by multiple sensors. Without requiring a priori probability of target presence, the fusion threshold bounds derived using Chebyshev's inequality ensure a higher hit rate and lower false alarm rate compared to the weighted averages of individual sensors. The Monte Carlo-based simulation results show that the proposed approach significantly improves target detection performance, and can also be used to guide the actual threshold selection in practical sensor network implementation under certain error rate constraints.

Automatic license plate recognition (ALPR) is a pattern recognition application of great importance for access, traffic surveillance and law enforcement. Therefore many studies are concentrated on creating new algorithms or improving their performance. Many authors have presented algorithms that are based on individual methods such as skeleton features, neural networks or template matching for recognizing the license plate symbols. In this paper we present a novel approach for decisional fusion of several recognition methods, as well as new classification features. The classification results are proven to be significantly better than those obtained for each method considered individually. For better results, syntax corrections are also considered. Several trainable and non-trainable decisional fusion rules have been taken into account, evidencing each of the classification methods at their best. Experimental results are shown, the results being very encouraging by obtaining a symbol good recognition rate (GRC) of more than 99.4% on a real license plate database.

The purpose of this paper is to present a view of a telemonitoring system based on fuzzy logic for distress situations detection of elderly people living alone in the home environment. This system includes three different sensors: physiological sensors (cardiac frequency, activity or agitation, posture and fall detection sensor), microphones and infrared sensors. Taking into account the difficulty of the statistical modeling of abnormal situation and considering that the Fuzzy Logic has been actually used with success in the development of classifier and analysis of control systems, its use in the decision fusion module of our home elderly people monitoring system is presented and evaluated in this paper.

A band clustering and selection approach based on a hyperspectral measure, spectral information divergence (SID) is presented in this paper. Hyperspectral image data is analyzed for target detection. Hyperspectral image data and spectral signatures of the targets are used to measure the SID. Virtual dimensionality (VD) is used to select optimal number of bands. For end member extraction, vertex component analysis (VCA) is used. For decision fusion a new approach based on spectral discriminatory entropy (SDE) is proposed. A comparative study is conducted to show the effectiveness of new approach of band clustering and selection. Decision fusion is also compared with full band and individual SID detection schemes.

A diagnostic tool for detecting damage to spiral bevel gears was developed. Two different monitoring technologies, oil debris analysis and vibration, were integrated using data fusion into a health monitoring system for detecting surface fatigue pitting damage on gears. This integrated system showed improved detection and decision-making capabilities as compared to using individual monitoring technologies. This diagnostic tool was evaluated by collecting vibration and oil debris data from fatigue tests performed in the NASA Glenn Spiral Bevel Gear Fatigue Rigs. Data was collected during experiments performed in this test rig when pitting damage occurred. Results show that combining the vibration and oil debris measurement technologies imprbves the detection of pitting damage on spiral bevel gears.

Today's wireless communication systems follow fixed spectrum assignment policies which leads to overall inefficient spectrum use. Further, spectrum scarcity is an issue for operators with emerging mobile services and large number of users with even higher capacity requirements. This inefficiency and scarcity in spectrum usage necessitates a new paradigm for communications such as utilising available spectrum opportunistically. Cognitive radio (CR) is an enabling technology having potential to increase spectrum utilisation and provide desired interference protection to licenced users. This can be done by detection of spectrum opportunities by secondary users. Due to channel fading and shadowing a single user can not make a reliable decision and collaboration of and among users is required. In this paper, it has been demonstrated that for improved detection performance decision fusion algorithm for collaborative spectrum sensing must have information about channel and the mean SNR of all secondary users. Using Monte Carlo simulations it is concluded that for optimum performance it is not necessary that all users collaborate with each other.

This paper demonstrates two different fusion techniques at two different levels of a human face recognition process. The first one is called data fusion at lower level and the second one is the decision fusion towards the end of the recognition process. At first a data fusion is applied on visual and corresponding thermal images to generate fused image. Data fusion is implemented in the wavelet domain after decomposing the images through Daubechies wavelet coefficients (db2). During the data fusion maximum of approximate and other three details coefficients are merged together. After that Principle Component Analysis (PCA) is applied over the fused coefficients and finally two different artificial neural networks namely Multilayer Perceptron(MLP) and Radial Basis Function(RBF) networks have been used separately to classify the images. After that, for decision fusion based decisions from both the classifiers are combined together using Bayesian formulation. For experiments, IRIS thermal/visible Face Database has been used. Experimental results show that the performance of multiple classifier system along with decision fusion works well over the single classifier system.

We consider Decentralized Detection (DD) of an unknown signal corrupted by zero-mean unimodal noise via Wireless Sensor Networks (WSNs). We assume the presence of both smart and dumb sensors: the former transmit unquantized measurements, while the latter employ multi-level quantizations (before transmission through binary symmetric channels) in order to cope with energy and/or bandwidth constraints. The data are received by a Fusion Center (FC), which relies on a proposed Rao test, as a simpler alternative to the Generalized Likelihood Ratio Test (GLRT). The asymptotic performance analysis of the multi-bit Rao test is provided and exploited to propose a (signal-independent) quantizer design approach by maximizing the non-centrality parameter of the test-statistic distribution. Since the latter is a non-linear and non-convex function of the quantization thresholds, we employ the particle swarm optimization algorithm for its maximization. Numerical results are provided to show the effectiveness of Rao test in comparison to GLRT and the boost in performance obtained by (multiple) threshold optimization. Asymptotic performance is also exploited to define detection gain measures allowing to assess gain arising from (a) use of dumb sensors and (b) increasing their quantization resolution.

Numerous detection algorithms, using various sensor modalities, have been developed for the detection of mines in cluttered and noisy backgrounds. The performance for each detection algorithm is typically reported in terms of the receiver operating characteristic (ROC), which is a plot of the probability of detection versus false alarm as a function of the threshold setting on the output decision variable of each algorithm. In this paper, we present multisensor decision-fusion algorithms that combine the local decisions of existing detection algorithms for different sensors. This offers an expedient, attractive, and much simpler alternative to the design of an algorithm that fuses multiple sensors at the data level, especially in cases of limited training data where it is difficult to make accurate estimates of multidimensional probability density functions. The goal of our multisensor decision-fusion approach is to exploit the complimentary strengths of existing multisensor algorithms so as to achieve performance (ROC) that exceeds the performance of any sensor algorithm operating in isolation. Our approach to multisensor decision fusion is based on the optimal signal detection theory using the likelihood ratio. We consider the optimal fusion of local decisions for two sensors: a groundpenetrating radar and a metal detector. A new robust algorithm for decision fusion that addresses the problem in which the statistics of the training data are not likely to exactly match the statistics of the test data is presented. ROCs are presented and compared for field data.

This paper describes a fusion of visual and thermal infrared (IR) images for robust face recognition. Two types of fusion methods are discussed: data fusion and decision fusion. Data fusion produces an illumination-invariant face image by adaptively integrating registered visual and thermal face images. Decision fusion combines matching scores of individual face recognition modules. In the data fusion process, eyeglasses, which block thermal energy, are detected from thermal images and replaced with an eye template. Three fusion-based face recognition techniques are implemented and tested: Data fusion of visual and thermal images (Df), Decision fusion with highest matching score (Fh), and Decision fusion with average matching score (Fa). A commercial face recognition software FaceIt ® is used as an individual recognition module. Comparison results show that fusion-based face recognition techniques outperformed individual visual and thermal face recognizers under illumination variations and facial expressions.

The technique of sensor fusion addresses the issues relating to the optimality of decision-making in the multiple-sensor framework. The advances in sensor fusion enable to perform intrusion detection for both rare and new attacks. This thesis discusses this assertion in detail, and describes the theoretical and experimental work done to show its validity.

In this paper, we address the multimodal biometric decision fusion problem. By exploring into the user-specific approach for learning and threshold setting, four possible paradigms for learning and decision making are investigated. Since each user requires a decision hyperplane specific to him in order to achieve good verification accuracy, those tedious iterative training methods like the neural network approach would not be suitable. We propose to use a model that requires only a single training step for this application. The four global and local learning and decision paradigms are then explored to observe their decision capability. Besides the proposal of a relevant receiver operating characteristic performance for the local decision, extensive experiments were conducted to observe the verification performance for fusion of two and three biometrics.

We study channel-aware decision fusion over a
multiple-input multiple-output (MIMO) channel in the large-
array regime at the decision-fusion center (DFC). Inhomogeneous
large-scale fading between the sensors and the DFC is consider
in addition to the small-scale fading, and pilot-based channel
estimation is performed at the DFC. Linear processing techniques
are analyzed in order to design low-complexity alternatives
to the optimum log-likelihood ratio test (LLRT). Performance
evaluation based on Monte Carlo simulations are presented.

Abstract. A new parallel hybrid decision fusion methodology is proposed. It is demonstrated that existing parallel multiple expert decision combination approaches can be divided into two broad categories based on the implicit decision emphasis implemented. The first category consists of methods implementing computationally intensive decision frameworks incorporating a priori information about the target task domain and the reliability of the participating experts, while the second category encompasses approaches implementing group consensus without assigning any importance to the reliability of the experts and ignoring other contextual information. The methodology proposed in this paper is a hybridisation of these two approaches and has shown significant performance enhancements in terms of higher overall recognition rates along with lower substitution rates. Detailed analysis using two different databases supports this claim.

Digital forensics, which identifies the characteristics and origin of a digital device, has become a new field of research. If digital content will serve as evidence in court, similar to its non-digital counterparts, digital forensics can play a crucial role in identifying the source model or device. To achieve this goal, the relationship between an image and its camera model will be explored. Various image-related and hardware-related features are utilized in the proposed model by a support vector machine approach along with decision fusion techniques. Furthermore, the optimum feature subset to achieve the highest accuracy rate is also explored.

Multichannel analysis is an innovative technique used for the analysis of bioelectrical signals. In this paper, we analyzed uterine Electromyogram (EMG) signals recorded by means of a 4x4 electrode matrix positioned on the woman's abdomen by using a multichannel approach. Relevant features were extracted from each channel and fed to a competitive neural network (CNN). First, we evaluated the classification performance of each channel. Then, we compared these performances to see which channel ranks better than the others. Finally, a decision fusion method based on the weighted sum of the individual decision of each channel was tested. The results showed that data can be grouped into 2 different groups. Furthermore, they showed that the classification performance varies according to the position of the electrode. Therefore, when a decision fusion rule was applied, the network yielded better classification accuracy than any individual channel could provide. These encouraging results prove that multichannel analysis can improve the classification of uterine EMG signals.

Abstract—In cognitive radio systems, secondary users can be coordinated to perform cooperative spectrum sensing so as to detect the primary user more accurately. However, when the sensing observations are forwarded to a common receiver through fading channels, the sensing ...

This article discusses the automatic recognition of Cued Speech in French based on hidden Markov models (HMMs). Cued Speech is a visual mode which, by using hand shapes in different positions and in combination with lip patterns of speech, makes all the sounds of a spoken language clearly understandable to deaf people. The aim of Cued Speech is to overcome the problems of lipreading and thus enable deaf children and adults to understand spoken language completely. In the current study, the authors demonstrate that visible gestures are as discriminant as audible orofacial gestures. Phoneme recognition and isolated word recognition experiments have been conducted using data from a normal-hearing cuer. The results obtained were very promising, and the study has been extended by applying the proposed methods to a deaf cuer. The achieved results have not shown any significant differences compared to automatic Cued Speech recognition in a normal-hearing subject. In automatic recognition of Cued Speech, lip shape and gesture recognition are required. Moreover, the integration of the two modalities is of great importance. In this study, lip shape component is fused with hand component to realize Cued Speech recognition. Using concatenative feature fusion and multi-stream HMM decision fusion, vowel recognition, consonant recognition, and isolated word recognition experiments have been conducted. For vowel recognition, an 87.6\% vowel accuracy was obtained showing a 61\% relative improvement compared to the sole use of lip shape parameters. In the case of consonant recognition, a 78.9\% accuracy was obtained showing a 56\% relative improvement compared to the use of lip shape only. In addition to vowel and consonant recognition, a complete phoneme recognition experiment using concatenated feature vectors and Gaussian mixture model (GMM) discrimination was conducted, obtaining a 74.4\% phoneme accuracy. Isolated word recognition experiments in both normal-hearing and deaf subjects were also conducted providing a word accuracy of 94.9\% and 89\%, respectively. The obtained results were compared with those obtained using audio signal, and comparable accuracies were observed.

In this paper, we propose a special fusion method for combining ensembles of base classifiers utilizing new neural networks in order to improve overall efficiency of classification. While ensembles are designed such that each classifier is trained independently while the decision fusion is performed as a final procedure, in this method, we would be interested in making the fusion process more adaptive and efficient. This new combiner, called Neural Network Kernel Least Mean Square 1 , attempts to fuse outputs of the ensembles of classifiers. The proposed Neural Network has some special properties such as Kernel abilities, Least Mean Square features, easy learning over variants of patterns and traditional neuron capabilities. Neural Network Kernel Least Mean Square is a special neuron which is trained with Kernel Least Mean Square properties. This new neuron is used as a classifiers combiner to fuse outputs of base neural network classifiers. Performance of this method is analyzed and compared with other fusion methods. The analysis represents higher performance of our new method as opposed to others.

In this contribution we introduce a novel approach to the combination of acoustic features and language information for a most robust automatic recognition of a speaker's emotion. Seven discrete emotional states are classified throughout the work. Firstly a model for the recognition of emotion by acoustic features is presented. The derived features of the signal-, pitch-, energy, and spectral contours are ranked by their quantitative contribution to the estimation of an emotion. Several different classification methods including linear classifiers, Gaussian Mixture Models, Neural Nets, and Support Vector Machines are compared by their performance within this task. Secondly an approach to emotion recognition by the spoken content is introduced applying Belief Network based spotting for emotional key-phrases. Finally the two information sources will be integrated in a soft decision fusion by using a Neural Net. The gain will be evaluated and compared to other advances. Two emotional speech corpora used for training and evaluation are described in detail and the results achieved applying the propagated novel advance to speaker emotion recognition are presented and discussed.

The main obstacle for a cognitive radio (CR) is to detect the presence of primary users (PUs) reliably in order to reduce the interference to licensed communications. Genetic algorithms (GAs) are well suited for CR optimization problems to improve spectrum utilization by manipulating its unused portions and offer a solution to the apparent spectrum underutilization problem. In this paper, we present binary genetic algorithm (BGA) and continuous genetic algorithm (CGA)-based soft decision fusion (SDF) scheme for cooperative spectrum optimization in cognitive radio network (CRN). Then

Efficient decision fusion strategies, for deriving optimal decisions in multisensor target recognitionltracking environments, are postulated and analyzed in this study. This fusion paradigm, which is designed for fusing decisions derived from a suite of parallel sensors, is embedded within a recursive system structure to achieve significant enhancement in the reliability of the fused decisions.

The purpose of this paper is to present a view of a telemonitoring system based on fuzzy logic for distress situations detection of elderly people living alone in the home environment. This system includes three different sensors: physiological sensors (cardiac frequency, activity or agitation, posture and fall detection sensor), microphones and infrared sensors. Taking into account the difficulty of the statistical modeling of abnormal situation and considering that the Fuzzy Logic has been actually used with success in the development of classifier and analysis of control systems, its use in the decision fusion module of our home elderly people monitoring system is presented and evaluated in this paper.

In this paper, we propose novel spectrally weighted mel-frequency cepstral coefficient (WMFCC) features for emotion recognition from speech. The idea is based on the fact that formant locations carry emotion-related information, and therefore critical spectral bands around formant locations can be emphasized during the calculation of MFCC features. The spectral weighting is derived from the normalized inverse harmonic mean function of the line spectral frequency (LSF) features, which are known to be localized around formant frequencies. The above approach can be considered as an early data fusion of spectral content and formant location information. We also investigate methods for late decision fusion of unimodal classifiers. We evaluate the proposed WMFCC features together with the standard spectral and prosody features using HMM based classifiers on the spontaneous FAU Aibo emotional speech corpus. The results show that unimodal classifiers with the WMFCC features perform significantly better than the classifiers with standard spectral features. Late decision fusion of classifiers provide further significant performance improvements.

There has been an increasing interest in the study of video based fire detection algorithms as video based surveillance systems become widely available for indoor and outdoor monitoring applications. A novel method explicitly developed for video based detection of wildfires at night (in the dark) is presented in this paper. The method comprises four sub-algorithms: (i) slow moving video object detection, (ii) bright region detection, (iii) detection of objects exhibiting periodic motion, and (iv) a sub-algorithm interpreting the motion of moving regions in video.

A robust model for the automatic segmentation of human brain images into anatomically defined regions across the human lifespan would be highly desirable, but such structural segmentations of brain MRI are challenging due to agerelated changes. We have developed a new method, based on established algorithms for automatic segmentation of young adults' brains. We used prior information from 30 anatomical atlases, which had been manually segmented into 83 anatomical structures. Target MRIs came from 80 subjects (~12 individuals/decade) from 20 to 90 years, with equal numbers of men, women; data from two different scanners (1.5T, 3T), using the IXI database. Each of the adult atlases was registered to each target MR image. By using additional information from segmentation into tissue classes (GM, WM and CSF) to initialise the warping based on label consistency similarity before feeding this into the previous normalised mutual information non-rigid registration, the registration became robust enough to accommodate atrophy and ventricular enlargement with age. The final segmentation was obtained by combination of the 30 propagated atlases using decision fusion. Kernel smoothing was used for modelling the structural volume changes with aging. Example linear correlation coefficients with age were, for lateral ventricular volume, r male =0.76, r female =0.58 and, for hippocampal volume, r male =-0.6, r female =-0.4 (all ρ<0.01).

Pansharpening is the process of improving spatial resolution of multispectral (MS) satellite images using the spatial details of a high resolution Panchromatic (PAN) image. Quality without reference (QNR) index is a tool which can be used to determine the global quality of the fused images without requiring a high resolution reference MS image. Over the years different pansharpening methods have been proposed using different criteria to obtain improved fusion quality. Since QNR can be used to determine the quality of fused images at full resolution without requiring a reference high resolution MS image, it may be used as a decision criterion to select the best fused regions, given a number of different fused images. In this paper we propose to use the QNR index to determine the local fusion quality. We also extend this property of local quality assessment to select regions among various fused images of the same region to obtain a single image with the best quantitative measures. The proposed decision fusion is tested on Pléiades and Quickbird satellite images fused by recently proposed fusion methods.

This paper presents four schemes for soft fusion of the outputs of multiple classi®ers. In the ®rst three approaches, the weights assigned to the classi®ers or groups of them are data dependent. The ®rst approach involves the calculation of fuzzy integrals. The second scheme performs weighted averaging with data-dependent weights. The third approach performs linear combination of the outputs of classi®ers via the BADD defuzzi®cation strategy. In the last scheme, the outputs of multiple classi®ers are combined using Zimmermann's compensatory operator. An empirical evaluation using widely accessible data sets substantiates the validity of the approaches with data-dependent weights, compared to various existing combination schemes of multiple classi®ers. Ó 0167-8655/99/$ ± see front matter Ó 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 -8 6 5 5 ( 9 9 ) 0 0 0 1 2 -4