No description available.

No description available.

CD-ROM in back pocket comprises experimental results and executables. / System requirements: Unix workstation or PC with Windows 95 or Windows NT. The reports output by EPrep. can be viewed with a web browser such as Netscape or Microsoft Internet Explorer through the top level HTML page. / Bibliography: p. 251-261. / Computer data and programs / HTML reports, data and figures generated by EPrep / xxiv, 261 p. : ill. ; 30 cm. + 1 computer laser optical disk ; 4 3/4". / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Proposes a framework for automatic feature extraction called generalised pre-processor. / Thesis (Ph.D.)--University of Adelaide, Dept. of Electrical and Electronic Engineering, 1999

CD-ROM in back pocket comprises experimental results and executables. / System requirements: Unix workstation or PC with Windows 95 or Windows NT. The reports output by EPrep. can be viewed with a web browser such as Netscape or Microsoft Internet Explorer through the top level HTML page. / Bibliography: p. 251-261. / Computer data and programs / HTML reports, data and figures generated by EPrep / xxiv, 261 p. : ill. ; 30 cm. + 1 computer laser optical disk ; 4 3/4". / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Proposes a framework for automatic feature extraction called generalised pre-processor. / Thesis (Ph.D.)--University of Adelaide, Dept. of Electrical and Electronic Engineering, 1999

The thesis illustrates a DSP-based speaker recognition system . In order to make the modular within the representation floating-point, we simplify the algorithm. This speaker recognition system is including hardware setting and implementation of speaker algorithm. The DSP chip is float arithmetic DSP(ADSP-21161 of ADI SHARK Series) , the algorithm of speaker recognition is gaussian mixture model. According to result of experiments, the speaker recognition of DSP can gain good recognition and speed efficiency.

No description available.

In this thesis, we proposed an innovation ideal that is employment of laser to extract finger feature, and constructed laser speckle recognition systems for this kind of feature. When projecting laser on the object surface, the speckle could be obtained to represent the characteristic of object surface by collecting scattered light. Two measurement of scattered light was adopted. First is laser signal recording the strength of scattered light when laser scan across the finger. The second is laser speckle image which is demonstrated when projecting the laser on the fingerprint and simultaneously collecting the scatter light by CCD. We proposed two recognition systems for laser signal and laser speckle. Besides, the proposed laser speckle fingerprint recognition system combines biometric detection, it can accurately distinguish biometric and non-biometric speckle. Experimental results demonstrate that proposed laser speckle recognition systems are feasible and with excellent ability of identity verification.

In this thesis, the speaker and emotion recognition system is established by PC and digit signal processor (DSP). Most speaker and emotion recognition systems are separately accomplished, but not combined together in the same system. In this thesis, it will show how speaker and emotion recognition systems are combined in the same system. In this system, the voice is picked up by a mike and through DSP to extract the characteristics. Then it passes the sample correctly, it can draw the result of distinguishing. The recognition system is divided into four sub-systems: the pronunciation pre-process, the speaker training model, the speaker and emotion recognition, and the speaker confirmation. The pronunciation pre-process uses the mike to capture the voice, and through the DSP board to convey the voice to the SRAM, then movements dealt with pre-process. The speaker trained model uses the Gaussian mixture model to establish the average, coefficient of variation and weight value of the person who sets up speaker specifically. And we¡¦ll take this information to be the datum of the whole recognition system. The speaker recognition mainly uses the density of probability to recognition the speaker¡¦s identity. The emotion recognition takes advantage of the coefficient of variation to recognize the emotion. The speaker confirms is set up to sure whether the user is the same speaker who hits for the systematic database. The recognition system based on DSP includes two parts¡GHardware setting and implementation of speaker algorithm. We use the fixed-arithmetician DSP chip (chipboard) in the DSP, the algorithm of recognition is Gaussian mixture model. In addition, compared with floating point, the fixed point DSP cost much less; it makes the system nearer to users.

This thesis presents an image-based real-time facial expression recognition system that is capable of recognizing basic facial expressions of several subjects simultaneously from a webcam. Our proposed methodology combines a supervised transfer learning strategy and a joint supervision method with a new supervision signal that is crucial for facial tasks. A convolutional neural network (CNN) model, MobileNet, that contains both accuracy and speed is deployed in both offline and real-time frameworks to enable fast and accurate real-time output. Evaluations for both offline and real-time experiments are provided in our work. The offline evaluation is carried out by first evaluating two publicly available datasets, JAFFE and CK+, and then presenting the results of the cross-dataset evaluation between these two datasets to verify the generalization ability of the proposed method. A comprehensive evaluation configuration for the CK+ dataset is given in this work, providing a baseline for a fair comparison. It reaches an accuracy of 95.24% on JAFFE dataset, and an accuracy of 96.92% on 6-class CK+ dataset which only contains the last frames of image sequences. The resulting average run-time cost for recognition in the real-time implementation is reported, which is approximately 3.57 ms/frame on an NVIDIA Quadro K4200 GPU. The results demonstrate that our proposed CNN-based framework for facial expression recognition, which does not require a massive preprocessing module, can not only achieve state-of-art accuracy on these two datasets but also perform the classification task much faster than a conventional machine learning methodology as a result of the lightweight structure of MobileNet.

An integrated heating, ventilation and air-conditioning (HVAC) system is one of the most important components to determining the energy consumption of the entire building. For commercial buildings, particularly office buildings and schools, the heating and cooling loads are largely dependent on the occupant behavioral patterns such as occupancy rates and their activities. Therefore, if HVAC systems can respond to dynamic occupancy profiles, there is a large potential to reduce energy consumption. However, currently, most of existing HVAC systems operate without the ability to adjust supply air rate accordingly in response to the dynamic profiles of occupants. Due to this inefficiency, much of the HVAC energy use is wasted, particularly when the conditioned spaces are unoccupied or under-occupied (less occupants than the intended design). The solution to this inefficiency is to control HVAC system based on dynamic occupant profiles. Motivated by this, the research provides a real-time vision-based occupant pattern recognition system for occupancy counting as well as activity level classification. The proposed vision-based system is integrated into the existing HVAC simulation model of a U.S. office building to investigate the level of energy savings as well as thermal comfort improvement compared to traditional existing HVAC control system. The research is divided into two parts. The first part is to use an open source library based on neural network for real-time occupant counting and background subtraction method for activity level classification with a common static RGB camera. The second part utilizes a DOE reference office building model with customized dynamic occupancy schedule, including the number of occupant schedule, activity schedule and clothing insulation schedule to identify the potential energy savings compared with conventional HVAC control system. The research results revealed that vision-based systems can detect occupants and classify activity level in real time with accuracy around 90% when there are not many occlusions. Additionally, the dynamic occupant schedules indeed can bring about energy savings. Details of vision-based system, methodology, simulation configurations and results will be presented in the paper as well as potential opportunities for use throughout multiple types of commercial buildings, specifically focused on office and educational institutes.