Recognition of Foreign Bodies for Grading of Food Grains using Machine Learning Techniques

Farmers sell food grains in market, which is usually assessed by visual inspection by food inspectors is a tedious and time consuming task. In this context, there is a dire need for development of automated system for grading of food grains. In this paper, work is carried out for recognition of foreign bodies (stone, stick, broken grain and worm) in food grain images which is supported by both window and android OS. The proposed mobile app recognizes foreign bodies in food grain images using Back Propagation Network (BPN). Outcome of the proposed work is displayed in the form of pie chart showing the relative percentage of foreign bodies (if any) and food grains in the input image. The developed automated system can be used by the robots for automated food grain assessment, automated sowing, wherein the robot can identify, grade, recognize foreign bodies in food grains.