ABSTRACT
The growth of aerobic bacteria on Korean seasoned soybean sprouts was modelled as a function of temperature to estimate microbial spoilage and shelf life on a real-time basis under dynamic storage conditions. Counts of aerobic bacteria on seasoned soybean sprouts stored at constant temperatures between 0 degrees C and 15 degrees C were recorded. The bootstrapping method was applied to generate many resampled data sets of mean microbial plate counts that were then used to estimate the parameters of the microbial growth model of Baranyi and Roberts. The distributions of the model parameters were quantified, and their temperature dependencies were expressed as mathematical functions. When the temperature functions of the parameters were incorporated into differential equations describing microbial growth, predictions of microbial growth under fluctuating temperature conditions were similar to observed microbial growth.
Subject(s)
Bacteria, Aerobic/growth & development , Food Handling/methods , Food Preservation/methods , Glycine max/microbiology , Models, Biological , Colony Count, Microbial , Consumer Product Safety , Kinetics , Korea , Temperature , Time FactorsABSTRACT
We demonstrate the operation of a novel all-optical flip-flop. The flip-flop consists of a slave Fabry-Perot laser diode (FP-LD) and a specially designed master FP-LD which has a built-in external cavity and operates in single longitudinal mode oscillation. The set and reset pulses were generated by external modulators at 1 Gbit/s. The rising and falling times of the output signal in on-off operation of the flip-flop were about 50 ps. The required powers of both set and reset pulses were less than -9 dBm.
ABSTRACT
TAG (Training by Adaptive Gain) is a new adaptive learning algorithm developed for optical implementation of large-scale artificial neural networks. For fully interconnected single-layer neural networks with N input and M output neurons TAG contains two different types of interconnections, i.e., M N global fixed interconnections and N + M adaptive gain controls. For two-dimensional input patterns the former may be achieved by multifacet holograms, and the latter by spatial light modulators (SLMs). For the same number of input and output neurons TAG requires much less adaptive elements, and provides a possibility for large-scale optical implementation at some sacrifice in performance as compared to the perceptron. The training algorithm is based on gradient descent and error backpropagation, and is easily extensible to multilayer architecture. Computer simulation demonstrates reasonable performance of TAG compared to perceptron performance. An electrooptical implementation of TAG is also proposed.
