In this study, we investigated the feasibility of constructing a peaberry identification system using visible and fluorescence images to automate the peaberry sorting process in coffee beans. Coffee beans were irradiated with a white LED and a 360 nm ultraviolet LED, respectively, and photographed with a color camera. The RGB values of fluorescence images were converted into HSV (Hue, Saturation, Value) values, and the comparison of each feature value showed that there was a significant difference in the saturation of the fluorescence images. The Peaberry test data were discriminated against using a convolutional neural network trained on visible and fluorescence images as input, and the Peaberry discrimination rate was more than 95%. Furthermore, to verify the effectiveness of the system that combines visible and fluorescence images for classification, we randomly picked up a total of 56 beans, 28 of which were misclassified by visible and fluorescence images, and judged the type again using the sum of the output values of both classifiers using visible and fluorescence images, and found that 72%, or 41, were classified correctly. That suggests that the use of multiple image information contributes to improving the accuracy of the peaberry identification system.

Tissue oxygenation evaluation is an essential element in the management and understanding of patient care. This evaluation requires measurement of hemoglobin concentration and its oxygen saturation in the blood. The current study proposed the principal equation for simultaneous optical measurement of hemoglobin concentration, oxygen saturation, and water percentage in the blood. In this equation, the absorption coefficient of hemoglobin for the light of wavelength λ (μa_λ) is expressed using the hemoglobin oxygen saturation (SO₂), hemoglobin concentration ([Hb]), water percentage in the blood (fw), molar extinction coefficient (ε) for the light of wavelength λ for oxygenated hemoglobin and deoxygenated hemoglobin, and the absorption coefficients of water (μa_λ^water) for the light of wavelength λ. Thereafter, a total of 125 arbitrary values of μa_λ were set by introducing various SO₂, [Hb], and fw values in the equation with 125 different combinations. We observed that all the combinations of SO₂, [Hb], and fw values could be reconstituted from the corresponding μa_λ values using the Nelder–Mead method, a heuristic optimization technique, with the principal equation. However, this was only possible when ε and μa_λ^water values were used for light of specific wavelengths, i.e., light wavelengths that were absorbed by oxygenated and deoxygenated hemoglobin and water. This result suggests that hemoglobin concentration, oxygen saturation, and water percentage in the blood may be estimated from the hemoglobin absorption coefficient values via the optimization method with the proposed principal equation.