Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 8 (2022) > Article

DOI: 10.14704/nq.2022.20.8.NQ44745

Dull Intensity Medicinal Plant Leaf Disease Detection using Machine Learning Algorithms

Meenakshi T, Dr.Pushpa Rani


For almost 70% population in India, agriculture is the primary source of income. Plants and their products are playing key role in human lives. Especially medicinal plants like Neem, Basil, Beal, Betal, Cardamom, Pepper, Garlic, Citrus, Clover,Aloe, Sandalwood, Turmeric etc., and are used in various ways like preparation of medicines, as a food species, herbal cosmetics and in holy rituals. Herbal or medicinal plant yielding and usage is growing rapidly now a days. The quality and yielding growth of crop is mainly depending on the diseases affected to the crop. Seasonal changes, climatic changes, soil types, lack of nutrition, lack of knowledge and proper timely usage of fertilizers, natural disasters and some others are the causes to the diseases of crops. Crop diseases are the main obstacles to get high quality and yielding. Huge losses or damage will occur, if diseases are not identified and controlled early. Manual detection of disease is time consuming, laborious and requires additional cost to the farmers. Automatic disease detection systems allow early identification of disease and save the farmer from huge losses. Captured leaf image is the key element in identification of using automated disease detection systems. Climatic conditions, sensor settings of the camera, noise and some other factors cause dull intensity or uneven illuminated intensity images. It is a challenging task to process and identification of diseases with these images. A new method is proposed using CLAHE and SVM classifier with modified kernel for disease detection and identification with dull intensity or uneven intensity illuminated leaf images. For preprocessing and enhancement of features in the dull intensity image, CLAHE and image processing preprocessing methods are used. GLCM is used for extraction of important features for classification. For identification and classification of disease type, SVM classifier with modified RBF kernel is used. The proposed method is providing good results than the existing methods. The performance metrics like accuracy, precision, recall and f1score are evaluated between proposed method and the existing method, proposed method is showing good performance in these metrics.


Automated disease detection system, crop diseases, CLAHE, dull intensity images, GLCM, RBF kernel, SVM classifier

Full Text