An Overview on Detection and Classification of Plant Diseases through CBIR for Mobile Application
- Select Volume / Issue:
- Type of Publication:
- CBIR, Hierarchical Clustering, Segmentation, Feature Extraction, SVM, Plant Diseases, Android Application
- Puchalwar, Saniya Santosh; Mangrulkar, Om Vivek; Bhoyar, Prajit Himdev; Lande, Milind V.
- An average farmer is unaware of the numerous illnesses that might infect his agricultural plantation. This might result in a decrease in agricultural output. Agriculture is one of the most important sources of income in developing countries. Soil quality, humidity, temperature, and disease all play an important part in crop production. Farmers are typically uninformed and unaware of environmental conditions that may impact crop cultivation. To assist farmers, we may use machine learning and image analysis tools to deliver useful information. This may be accomplished using a web portal and mobile phone We suggested in this work to construct a system that offers real-time feedback based on plant input photos. The use of technology has become critical in assisting farmers in gathering significant and up-to-date information and knowledge, which are key resources on which farming depends. The goal is to treat plant illnesses and so manage them by meticulously identifying plant leaves. The aim is to create a prototype that is timely, relevant, accurate, and simple to use update to account for fast technological change. Based on a query image, the Content Based Image Retrieval (CBIR) approach is used to recover photos of damaged plants from a training dataset. The obtained photos are segmented using Hierarchical Clustering, resulting in clusters of sick plant images. Support Vector Machine is then used to classify the clusters (SVM) Classifier based on cluster characteristics that checks the proper type of illness impacting the plant set.
Full text: IJRAS_1090_FINAL.pdf