Skip to main content
ARS Home » Plains Area » Mandan, North Dakota » Northern Great Plains Research Laboratory » Research » Publications at this Location » Publication #305388
Title: Machine vision analysis for industrial beet color change kinetics and total soluble solid content

Author
item POTHULA, ANAND - North Dakota State University
item IGATHINATHANE, CANNAYEN - North Dakota State University
item SHEN, JIACHENG - North Dakota State University
item Nichols, Kristine
item Archer, David

Submitted to: Proceedings of the American Society of Agricultural and Biological Engineers International (ASABE)
Publication Type: Proceedings
Publication Acceptance Date: 3/28/2014
Publication Date: 5/1/2014
Citation: Pothula, A.K., Igathinathane, C., Shen, J., Nichols, K.A., Archer, D.W. 2014. Machine vision analysis for industrial beet color change kinetics and total soluble solid content. Proceedings of the American Society of Agricultural and Biological Engineers International (ASABE). Intersectional Meeting, Paper No. SD14-008, p. 1-9.

Interpretive Summary: Industrial beets are being developed as a crop for use in production of bioenergy and other bioproducts. An automated process is needed to rapidly identify beet quality, including the amount of sugars contained in the beets as they are being processed. Recognizing that the color of crushed beets rapidly changes after crushing, and that the rate of color change is related to the sugar content, a system was developed to measure the rate of color change and use the measured rate to estimate sugar content. The system uses a digital camera to measure the color at measured times after crushing. A mathematical model was developed based on the rate of color change over time. The best model provided an excellent fit for the data and the modeled color change measure provided a linear estimate of sugar content. The results demonstrate a technique that could be used to rapidly assess beet quality. The research findings could be used to develop a sensor for use in a processing facility to monitor beet quality as it is being processed. This could help reduce processing costs and maximize biofuel and bioproduct production from industrial beets. This research is important to commercial beet processors and the biofuel and bioproducts industry.

Technical Abstract: A machine vision system (MVS) for the measurement of color change kinetics in crushed industrial beet to evaluate the total soluble solid content (°Brix) was developed in this study. It is expected that higher the °Brix faster the color change and modeling this color change kinetics helps in assessing the ground beet quality for juice extraction. The central portion from the whole beet was chopped off for preparing the ground sample. Laboratory blender was used to grind the beets. Five different concentration of beet samples were prepared by mixing the crushed beets with varying quantities of cold water followed by draining. Digital images at regular time intervals were acquired for all the samples using a digital camera with an auto timer setting under a constant lighting condition. MVS was calibrated using the ColorChecker before the experimentation. Quadratic model was used for converting the RGB to L*,a*,b* values. For each sample a constant representative window of the image were cropped and analyzed for L*, a*, b* (CIELAB color space) and total color change. Program for calibrating, cropping and analyzing the images was written in MATLAB. Different color kinetics models were fitted separately for samples of different soluble solid concentration. Out of the models tested, the Page model gave the best fit (R2 = 0.91-0.99) for the color kinetics. Out of the model constants, Page model’s constant (n) gave an excellent linear fit (R2 = 0.995) with TSS of the crushed beets. The developed machine vision system can be successfully used for modeling the color change kinetics as well as measuring the beet sugar content by correlation that is mostly represented by the TSS.