Presentation Information
[16a-M_135-5]Study on Wavelength-Selective Filters for Agricultural Crop Evaluation Based on Surface Color
〇(M2)Sou Takamatsu1, Hiroshi Murotani1 (1.Tokai Univ)
Keywords:
light and color
This study investigates the effectiveness of selective wavelength filtering for improving human visual discrimination in the appearance evaluation of agricultural products. Tomatoes were selected as the target material, and spectral reflectance data from multiple sample pairs were analyzed through numerical simulations. Color differences (ΔE) were calculated under a no-filter condition and under various band-pass filter conditions with different center wavelengths and bandwidths. As an evaluation criterion, the minimum ΔE among all sample pairs was used to identify filter conditions that ensure discriminability across all samples.The results indicate that a band-pass filter with a center wavelength of 500 nm and a bandwidth of 40 nm effectively enhances color differences in visually difficult sample pairs while maintaining the discrimination threshold for visually easy pairs. This suggests that the selected filter condition improves perceptual contrast where discrimination is challenging, without compromising overall visual reliability. The observed changes in color difference are attributed to the relative enhancement of wavelength components that contribute to discrimination, accompanied by the suppression of redundant spectral information.These findings provide a quantitative guideline for designing selective wavelength-filtering eyewear to support human visual inspection. The proposed approach has potential applications in agricultural quality assessment, such as ripeness evaluation and appearance-based sorting, where intuitive and reliable visual judgment is required.