Comparison of RGB and white LED illumination in honey imaging system using histogram approach

LED illumination-based multispectral imaging (LEDMSI) is one of the effective methods in the spectral image acquisition system. The availability of high-intensity LEDs with many different colours makes it possible to produce satisfying data for multivariate image analysis, including deep learning multispectral systems. However, the impact of white LED correlated colour temperature (CCT) on data images has received attention in the field. In this paper, we aimed to compare the effect of red-green-blue (RGB) and CCT of an LED light source on honey images. Light and dark honey are considered image samples in the experiment. The honey images were recorded under the influence of RGB LED and three categories of CCT white LED using an 18-megapixel DSLR camera. All the pictures are saved in a JPEG format and uploaded in ImageJ software to acquire the spectra measurement (raw data) and histogram analysis. The result showed that CCT variations indicated a variation histogram as shown in the variation of RGB LED. Moreover, grayscale intensity analysis shows better brightness, contrast, and distinction between the two types of honey.