In this paper, we demonstrate how an existing deep learning architecture for automatically lip reading individuals can be adapted it so that it can be made speaker independent, and by doing so, improved accuracies can be achieved on a variety of different speakers. The architecture itself is multi-layered consisting of a convolutional neural network, but if we are to apply an initial edge detection-based stage to pre-process the image inputs so that only the contours are required, the architecture can be made to be less speaker favourable. The neural network architecture achieves good accuracy rates when trained and tested on some of the same speakers in the ”overlapped speakers” phase of simulations, where word error rates of just 1.3% and 0.4% are achieved when applied to two individual speakers respectively, as well as character error rates of 0.6% and 0.3%. The ”unseen speakers” phase fails to achieve as good an accuracy, with greater recorded word error rates of 20.6% and 17.0% when tested on the two speakers with character error rates of 11.5% and 8.3%. The variation in size and colour of different people’s lips will result in different outputs at the convolution layer of a convolutional neural network as the output depends on the pixel intensity of the red, green and blue channels of an input image so a convolutional neural network will naturally favour the observations of the individual whom the network was tested on. This paper proposes an initial ”contour mapping stage” which makes all inputs uniform so that the system can be speaker independent.
We present a new scheme for the inverse analysis of opto- thermal transients that is able to calculate initial temperature profiles from noisy data, despite the ill- conditioning. It is particularly useful for tackling problems where the inhomogeneities of the sample are insufficiently well understood, such as the diffusion of externally applied substances through skin or skin pigment depth profiling. We present the mathematical approach, supported with examples of analyses of in-vivo opto-thermal measurements.
Opto-thermal in-vivo measurements of stratum corneum hydration can now be evaluated to give hydration depth profiles from a mathematical model that incorporates both the surface hydration and the changes of hydration with depth beneath the surface. We now present an analysis of water diffusion in the stratum corneum that can be related to such measurements. The results are compared with measurements of changes of surface hydration and hydration gradient with depth in a stratum corneum stripping experiment. Surface hydration and hydration gradient information can also be combined to yield a measure of stratum corneum thickness, as illustrated with a measured thickness map of the volar forearm.
We report the results of a study aimed at comparing Opto-Thermal Transient Emission Radiometry (OTTER) with established techniques of assessing skin condition, namely evaporimetry (TEWL), skin dielectric constant measurement, ATR-FTIR and clinical assessment. Comparisons were made during a week-long study of the effects of intensive washing on the volar forearms of 14 subjects. The study also provided a comparison of skin condition after washing with two different cleansers, a mild isethionate betaine cleansing bar and a soap bar. The subject-averaged results from OTTER and TEWL were found to correlate with the clinical assessments, namely that intensive washing with the soap bar produces greater skin damage than with the isethionate betaine bar. Skin dielectric constant measurements were found to be sensitive to changes of skin condition other than hydration, as evidenced by a daily oscillation that dominate the results. The ATR-FTIR measurements proved difficult to evaluate, because of interfering calcium deposits from the soap bar. On the practical side, OTTER and skin dielectric constant measurements were found to be quicker and more convenient to use than TEWL and ATR-FTIR.
Opto-Thermal Transient Emission Radiometry has been shown to be able to measure stratum corneum hydration in-vivo. We now present a study that shows how the technique can be extended to give information about the distribution of water within the stratum corneum. This depth profile measurement capability is based on a new theoretical expression for the opto- thermal signal that takes a gradient of water concentration explicitly into account. The technique is illustrated with measurements on untreated skin and skin treated with petroleum jelly and DMSO.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.