Intelligent measurement of the color of food products

摘要

Abstract: Food color is an important attribute of food quality. This paper discusses issues in using machine vision for the objective assessment of food color, such as color image segmentation, calibration and statistical parameter design, and their application to pea color assessment. The algorithm first established a 3D histogram on RGB color space based on an octree data structure. From the color space, a statistical index was calculated to measure the major color of an object simulating the characteristics of human color perception. This method was applied to assess the major color of peas. Accuracy and efficiency were achieved in measuring pea seed and its variations. !15