Prediction of tannin, protein, and total phenolic content of grain sorghum using image analysis and machine learning

Abstract

Background and Objectives: Sorghum is an alternative crop where poor soil is a limiting factor for the production of corn. Laboratory measurements of chemical compounds of grains are expensive, time-consuming predominantly done on powdered samples. Therefore, a rapid and reliable method integrating image processing and machine learning was evaluated for the prediction of total phenolic compounds (TPCs), tannin, and protein.Findings: The highest TPC (0.83%) was obtained for genotype KGS36 (the brown pericarp) followed by KGS23 (0.54%). The concentration of tannin ranged from 0.008% to 0.616%. Mean comparison revealed variation for the protein concentration (15.17%-11.53%); the highest content obtained by KGS25 and the lowest by KGS36 (p < .01). Multilayer perceptron (MLP) as one of the common artificial neural networks (ANNs) applied to simulate the chemical of the grains was evaluated through the textural features extracted from grain images. For the MLP network, the minimum number of hidden networks was set to 8 and the maximum number was set to 25. For learning and saving, the number of networks was set to 20 and the number of saved networks was set to 5. The best MLP models were selected based on performance (R) and error values for the train, test, and validation sets. Conclusions: An ANN for the prediction of the chemical concentration was suggested. The correlations between predicted and observed data were higher than .915.Significance and Novelty: These models are of great significance for the prediction of chemical concentrations of grain sorghum in plant breeding and food industry.

Application of Imaging and Artificial Intelligence for Quality Monitoring of Stored Black Currant (Ribes nigrum L.)

“…In further studies, models to predict the chemical properties of stored black currants can be developed. The available literature data indicate the possibility of using image processing and machine learning to estimate the chemical properties of plant materials [ 44 , 45 ].…”

Evaluating the Classification of Freeze-Dried Slices and Cubes of Red-Fleshed Apple Genotypes Using Image Textures, Color Parameters, and Machine Learning

“…Image analysis and machine learning were successfully applied in previous studies to classify fruits and vegetables [17][18][19] and detect changes in the product quality as a result of different processing, such as drying and fermentation [20][21][22][23]. Furthermore, image features were used to estimate and predict the chemical properties of food products [24,25]. Machine-learning techniques can be very effective in predicting chemical substances [26].…”