International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
Farmer’s Tech Toolbox-Crop Recommendation and Disease Detection Saloni Raorane1, Kirti Singh2, Sakshi Tawte3, Vidhi Shiste4, Jyoti Bansode5, Ashvini Ahirrao6, 1Shah and Anchor Kutchhi Engineering College, Maharashtra, India
2Shah and Anchor Kutchhi Engineering College, Maharashtra, India 3Shah and Anchor Kutchhi Engineering College, Maharashtra, India 4Shah and Anchor Kutchhi Engineering College, Maharashtra, India
5Assistant Professor, Dept. of Information Technology, Shah and Anchor Kutchhi Engineering College,
Maharashtra, India
6Assistant Professor, Dept. of Information Technology, Shah and Anchor Kutchhi Engineering College,
Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Agriculture farming is an important sector for
knowledge, which may not always be accurate or take more time to detect the emerging threats like plant diseases. In Crop Recommendation, various factors such as temperature, soil nutrient levels (nitrogen, phosphorus, potassium), humidity, and pH are considered to assist farmers in selecting suitable crops for specific land plots. Utilizing advanced recommendation tools, farmers can anticipate the most appropriate crops to cultivate based on these environmental parameters. In Plant Disease Detection, farmers can upload images of their plants to identify potential diseases and monitor their plant health. This method enables farmers to promptly diagnose plant diseases and track the types of diseases affecting their crops. Furthermore, the scalability and adaptability of the proposed system hold the potential to drive sustainable agricultural practices and enhance food security on a global scale.
each country's growth. Many technologies like ML and DL are being implemented for farmers for the growth of crops. This research paper presents a novel approach to address this challenge by integrating crop recommendation with plant disease detection. Leveraging machine learning and image processing techniques, our system aims to provide farmers with intelligent decision support for crop selection and disease management. The methodology involves the collection of extensive datasets comprising crop information, soil conditions, climate data, and images of plant diseases. Deep learning models like RESNET models are employed for accurate disease detection from these images, while decision support algorithms analyze multiple factors to recommend suitable crops based on local conditions and disease prevalence. The results of our study demonstrate promising accuracy in disease detection and effectiveness in crop recommendation, thus offering a practical solution to enhance agricultural productivity and sustainability. This research contributes to the advancement of precision agriculture and underscores the potential of technology-driven approaches to address critical challenges in food security and agricultural sustainability.
2. RELATED WORK In,"Crop Recommender System Using Machine Learning Approach", it includes 76-90% accuracy in assessing weather impacts on crops in Madhya Pradesh. Another system compared supervised and unsupervised learning methods to predict yields and recommend fertilizers based on soil quality. Decision tree classifiers and random forest models have also been explored, alongside clustering techniques for improved accuracy. The current paper builds on these previous systems to advance crop yield prediction further [1]. In "Improving Crop Productivity Through A Crop Recommendation System Using Ensembling Technique" (Kulkarni et al., 2018), an ensemble model is developed to recommend optimal crops based on soil data. By combining multiple models, such as Random Forest and Naive Bayes, the system achieves 99.91% accuracy in classifying crops for Kharif versus Rabi growing seasons. This approach aims to boost agricultural productivity by offering tailored guidance to farmers regarding suitable crops for their soil conditions. Further advancements in predictive farming systems hold promise for benefiting farmers and the wider economy [2]. In "Prediction of Crop Yield and Fertilizer Recommendation Using Machine Learning Algorithms" (Bondre &
Key Words: Deep Learning, Random Forest, Logistic Regression, Feature Extraction, Accuracy.
1.INTRODUCTION Agriculture is the foundation of global food security, but it faces significant difficulties such as changing environmental conditions and the ongoing danger of plant diseases. In recent years, the use of technology into agricultural methods has shown considerable promise for tackling these difficulties. One such novel strategy is the combination of crop recommendation systems and plant disease detection methods. Using machine learning, image processing, and data analytics, this study aims to create a holistic system that provides farmers with intelligent decision support tools for crop selection and disease control. Traditionally, agricultural practices rely on subjective assessments and historical
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 2708