Qualitative Wheat Age Testing with Relation to Amylose Content Through Image Analysis-Based Web Algorithm

Prolonged storage of wheat can lead to a reduction in hydrolyzable starch content, rendering the wheat less digestible. Thus, wheat age, a crucial parameter to classify its quality, is significant for revenue generation through export. In this study, an image-based qualitative analysis of starch content has resonated with the aging of wheat. Certified wheat samples of different timelines have been treated with iodine solution in a 96 well plate to generate the bluish-black color of the starch-iodine complex. The well plate was then scanned with our previously reported UIISScan 1.1, a high-throughput field-portable sensory device based on advanced imaging array technology, and the image was analyzed with a unique web-based algorithm to correlate output colors of different years samples with their respective age. Various image features were analyzed using a machine learning algorithm to generate a significant color index value. Index values of different aged wheat samples were then investigated to develop a wheat age prediction model. This endeavor reports an overall 87% accuracy in confirming wheat age through the prediction model. This is the first ever reported web application-based image analysis model for qualitative wheat age prediction to the best of our knowledge.