Classification of Grain s and Quality Analysis u sing Deep Learning

There are various varieties of Rice and lentils. Price fabrication adulteration have been some the issues faced by consumers, farmers wholesale retailers. Traditional methods for Identification these similar types grains their quality analysis crude inaccurate. Methods were tried to implemented earlier but due financial inability low efficiency, they weren’t successful. To overcome this problem, project proposes a method that uses machine learning technique identification grains. Lentils which maximum consumption selected. designated into classes based on colors. The determining elegance lentil is with aid seed coat shade. Red lentils can be confirmed through cotyledon coloration. Lentil may also huge variety colors from inexperienced, red, speckled black tan. colour yellow or inexperienced. size color every Indian type (i.e. Red, Green, Yellow, Black, White) decided large Medium small, then end up part grade name. An smart used perceive kind bulk samples. proposed allows kernel length coloration using picture processing techniques. These measurements, when combined attributes sample, classify three commonly grown in India highest accuracy. one most consumed so its utmost importance. In project, we identify five rice them help distinguished features such as size, color, shape, surface. works phases viz., Feature Extraction, Training, Testing. Various grain has different surface come colors, Hence feature will extracted texture regression adopted grading mechanism where output terms percentage purity. methodology extraction GLCM Edge Detection supervised SVM Back Propagation utilized. provides an efficient replacement traditional standardizes pricing farm products only.