Macro-Nutrient Deficiency Identification in Plants Using Image Processing and Machine Learning

Plants need adequate amount of nutrients for complete and healthy life-cycle. Sufficient amount of six macronutrients like Nitrogen, Calcium, Phosphorous, Potassium, Sulphur, Magnesium are more important for the normal and healthy growth of plants. Lack or absence of nutrients leads to difficulties in carrying in the daily activities of plants and ultimately affect the yield. Thus, having the timely check for nutrient contents is important. Plants usually show the definite deficiency on their leaves with notable different patterns for each nutrient. Our proposed work is to provide an automated and reliable economic solution for nutrient deficiency identification. The dataset for deficient leaves and healthy leaves are created using image processing approach for RGB color feature extraction, real-time texture detection, edge detection, etc. This created dataset will be given to supervised machine learning as training dataset for further detection and identification of exact nutrient deficiency and healthy plants in order to take preventive measures to maximize the yield.