The part Ⅰ was for developing regression models to predict the moisture content, protein content and viscosity of brown and milled rice using Near Unfrared (NIR) Reflectance analyzer. The purpose of this study(part Ⅱ) is to measure fundamental data required for the prediction of rice quality , and to develop regression models to predict the protein content of brown and milled rice, brown rice yield from undreid paddy powder by using Near Infrared (NIR) Reflectance analyzer. The results of this study were summarized as follows . The predicted values of protein contents obtained from the undried paddy powder were will correlated to the measured values from brown and milled rice. The predicted yields of brown rice from undried paddy powder were not well correlated to be lab measured values from dried paddy. Continuous study in wavelength selection and of constituent relationship is necessary for practical application.

The structural characteristics of small scale rice polisher was analyzed to improve its performance. Spraying characteristic of nozzles used for rice polishing was also analyzed by a machine vision system. The internal pressure of the polishing chamber was measured according to outlet resistance, water spraying , and roller shaft speed. In addition , the performance of the rice polisher was evaluated to improve it in the basis of internal pressure in polishing chamber, whiteness , and broken rice ratio of clean rice according to the operating conditions. Actual nozzle discharge rate and drop size were 125cc/min and 86.97㎛, respectively. In the case of water spraying on rices, the internal pressure showed 4.9-9.8N/㎠ increase, broken rice ration decreased , and there was no difference in whiteness . The internal pressure increased up to two time with the increase of the outlet resistance. Also, the pressure at the upper part of screen was one and half times as high as the pressure at the lower part. In the case of water spraying rate of 150 cc/min, the roller shaft speed of 850 rpm resulted in no difference in whiteness and decrease of 0.3%in broken rice ratio, comparing to the roller shaft speed of 950 rpm.

It was assumed that the voice of cattle is generated with the virtual white noise through the digital filter called the linear prediction filter, and filter parameters (prediction coefficients) were estimated by the maximum entropy method (MEM) , using the sound signal of the animal . The feature planes were defined by the pairs of two parameters selected appropriately from these parameters. The cattle voices were divided into three levels, that is the high, medium and low levels according to their total power equivalent to the variances of the sound signal . It was found that the straight lines could be used for recognizing tow cow and one calf for high level voices. For high and medium level voices, however, it was difficult or impossible to recognize individual cattle on the parameters planes.