Some people often suffer from periodontal diseases and poor oral environments can be the cause of such diseases. The maintenance of oral health is a crucial issue for general health. One of the measures to maintain oral health is by brushing teeth to remove plaque. However, it is difficult to remove plaque perfectly because the color of plaque is similar to the color of the teeth making it difficult to find remaining plaque by simple visual inspection. In this study, we propose a functional illumination that helps detecting plaque on teeth. The functional illumination was optimized to enhance the color difference between the plaque and the teeth in a CIELAB color space. We designed the spectrum of the illumination by NLP (Non Linear Programming) in such a way that the color difference between the teeth and the plaque is maximized. The optimized illumination has a peak at 420 nm because the largest difference is observed in the reflectance between the teeth and the plaque in this wavelength. The color difference increased with the functional illumination by a factor of 3 with respect to a normal white light source. Additionally, we implemented the real functional illumination by a combination of LEDs and it was able to enhance color discrimination. For different participants, the functional illumination could also enhance color discrimination.