Rice grains with chalkiness will affect the quality and lower the price of rice, affecting the competitiveness of rice-exporting countries, and Vietnam is no exception. However, rice quality inspection is done manually, which takes time and money. The research proposes using image processing methods to determine the chalkiness, average width and percentage of whole grains (without broken) of each type of rice by image processing methods, especially the determination of grain grade. The level of chalkiness will be initially assessed according to Vietnamese standards. The study used 10 rice samples. The number of samples corresponding to 10 different types of rice that had been manually tested in the laboratory by humans was used as data to analyze the above indicators. The image processing methods are applied to analyze the color difference between the rice grain's chalkiness and the clear white area. Measured quality indicators are compared with methods in the laboratory by humans, and the reliability is much higher. This result facilitates the development of an assessment and verification tool to replace manual methods