Search

This chapter starts with a general and brief introduction of biometrics and audiovisual person recognition using mobile phone data. It begins with a discussion of what constitutes a biometric recognition system, and it then details the steps followed when audio-visual signals are used as inputs. This is followed by a review of the existing speaker and face recognition systems which have been evaluated on a mobile biometric database. We then discuss the key motivations of using deep neural network (DNN) for person recognition. We finally introduce a Deep Boltzmann Machine (DBM)- DNN, in short DBM-DNN, based framework for person recognition. An overview of the sections and sub-sections of this chapter is shown in Figure 4.1.

This paper presents a confidence-based late fusion framework and its application to audio-visual biometric identification. We assign each biometric matcher a confidence value calculated from the matching scores it produces. Then a transformation of the matching scores is performed using a novel confidence-ratio (C-ratio) i.e., the ratio of a matcher confidence obtained at the test phase to the corresponding matcher confidence obtained at the training phase. We also propose modifications to the highest rank and Borda count rank fusion rules to incorporate the matcher confidence. We demonstrate by experiments that our proposed confidence-based fusion framework is more robust compared to the state-of-the-art late (score- and rank-level) fusion approaches.

The mechanical properties of cycled aggregate concrete were studied. The recycled aggregates were prepared from the old concrete structures. It was observed that the recycled aggregates are covered with loose particles preventing a good bonding between the new cement matrixes and the cycled aggregates. The old cement paste that remained on the natural aggregate was porous and cracked leading to weak mechanical properties of the recycled aggregates. In the present study, a treatment by impregnation of ordinary Portland cement solution to the recycled aggregates was applied to improve their mechanical properties. The compressive strength tests of recycled aggregates concrete were carried out at two ages of curing for two target strengths. The test results showed that the compressive strengths of recycled aggregate concretes with the given treatment increase by 21 and 14%, respectively for 7 and 27days of curing, in compared to the normal aggregate concrete for the target strength of 3000 psi while for the target strength of 4000 psi, it was found 19 % and 13 %, respectively.

This paper presents a reliability estimation technique for an Audio-Visual (AV) person identification system. The estimated reliability measures are mapped to modality weights by using a mapping function. The weighted sum rule of fusion was used to fuse the audio and visual modalities before reaching a decision about the identity of a person. The proposed technique was tested on 88 subjects from the AusTalk AV database. Experimental results show that the proposed technique improves the identification accuracy of the min-max normalization, compared to when it is used without any reliability estimation. The identification accuracy of the proposed technique is also higher than the non-learned approach of modality weight adaptation.

Substrate clogging is one of the worst operational problems in horizontal sub-surface flow constructed wetlands (HSSFCWs). Clogging affects the treatment efficiency and life span of the system. This study presents the comparison of solids removal and clogging progress in aerated and non-aerated HSSFCWs. Experiments were conducted during 16 weeks in four plastic containers (0.24m2) filled with 38 cm of gravel. S1NON-AER was a normal non-aerated HSSFCW with plants (Phragmites australis). S2AER was also planted with Phragmites and was aerated by means of compressed air. S3HYBR was a hybrid CW, consisting of S3aAER and S3bNON-AER in series. S3aAER was similar to S2AER except that it only had half gravel depth. S3bNON-AER had equal gravel depth as S3aAER but was not planted and not aerated. In this hybrid system a part of the influent was bypassed directly to S3bNON-AER as carbon source for denitrification. During the experimental period, the aeration rate, influent concentration and bypass were manipulated. Solids (TSS & VSS) removal efficiency, hydraulic conductivity (HC), biofilm density, micro-invertebrate population, microbial activity (FDA hydrolysis) and fate of accumulated solids in each system were determined. Overall TSS removal efficiency was 77% for S1NON-AER, 93% for S2AER and 91% for S3HYBR. VSS removal efficiency was 81% for S1NON-AER, 96% for S2AER and 94% for S3HYBR. This indicates that solid removal efficiency is significantly higher in aerobic systems compared to anaerobic systems. HC was also considerably higher in aerobic systems (214 md-1for S2AER and 188 md-1for S3aAER) compared to anaerobic systems (105 md-1for S1NON-AER and 131 md-1for S3bNON-AER). Biofilm density (measured with protein, TOC and polysaccharides as proxies) was significantly higher near the inlet of anaerobic systems than that of aerobic systems. Biofilm density near the inlet and HC were negatively correlated. Ciliates were the dominant group among micro-invertebrate population.