Classification of Nutritional Deficiencies in Cabbage Leave Using Random Forest

Now a day agriculture is very important in India since it is a growing nation. But generally the crop production attained by farmers would be much below the optimal production. It is very important to correctly detecting and identifying the crop diseases to enhance the profit of the formers and the stakeholder. The main reason for the crop production gap is due to the lack of essential soil nutrients and irrigation in the agricultural farms. To escalate the crop production, it is essential to balance the chemical elements or nutrients present in the soil with varying parameters of soil like the pH and soil moisture. Crop productivity can be increased to optimum level by efficient soil nutrient management. In case of Nutrient deficiencies, visual symptoms will appear on the leaf. This paper put forwards a method to identify the nutrient deficiencies of plants by making use of visual symptoms appearing on the leaves by Classification. Eight types of deficiencies i.e. N, P, K, Ca, B, Zn and Mg will be studied. The proposed study consists of creation and pre¬processing of a set of images consisting of nutrient deficient and healthy leaves, feature extraction and by using Random Forest performing multi class classification of nutrient deficient leaves. Evaluation of tomato leaf from the dataset focuses on recognizing the visual detection and indications of nutritional deficiencies. The proposed architecture achieves the 98.30% accuracy with the model size of 9.20 MB.