International Research Journal of Engineering and Technology (IRJET) Volume: 11 Issue: 02 | Feb 2024 www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
REVIEW PAPER ON SMART CROP, FERTILIZER RECOMMENDATION AND PLANT DISEASE DETECTION Ayush Moundekar1, Harshal Mohadikar2, Kishor Thakre3 & Sarang Shivhare4, Prof. N. R. Hatwar 5 & Prof. R. V. Kurhekar 6 1234 Undergraduate Student, Department of Information Technology, Priyadarshini College of Engineering Nagpur,
Maharashtra India
56 Professor, Department of Information Technology, Priyadarshini College of Engineering Nagpur, Maharashtra India
-----------------------------------------------------------------------***-----------------------------------------------------------------Abstract - The foundation of our nation is the agricultural Keywords - Machine Learning, Crop & Fertilizer sector. In order to make farming simpler and to increase productivity, several new technologies are being incorporated into the field, including Deep Learning and Machine Learning. Farmers often choose the wrong crops, planting them in the wrong season or choosing ones that wouldn't produce much for the specific soil. A reduced yield will always be the consequence of improper crop selection. We created a machine learning recommendation model that describes the ideal crop to grow and the seed fertilizer depending on the weather & Soil conditions. Third-party apps that service APIs for climate and temperature, soil category, soil nutritious content, quantity of rainfall in that zone, and soil configuration are connected with location. It also kinds use of nutrient percentages for nitrogen (N), phosphorus (P), and potassium (K). When creating a model from the data, all of these features will be looked at utilizing a variety of legitimate machine learning approaches.
Recommendation & Plant Disease Detection, Decision Trees & XG Boost Algorithm
1. Introduction Approximately 58% of the workers in our nation is working in farming, making it one of the most important foundations of income. The majority of India's 1.2 billion workers work in agriculture. India is the world's secondlargest producer of fruits and vegetables; nevertheless, crop loss puts farmers in jeopardy, according to the Ministry of Agriculture, Co-operation and Farmer Welfare's annual report. Modern agricultural techniques are being implemented for the benefit of farmers. To greatly rise productivity and value, Deep learning, machine learning, and data mining methods are now being used by a few researchers. The state of the soil, which in turn depends on the nutrients in the soil, is the primary factor influencing agricultural production. Farmers should be advised on crops based on soil analysis in order to boost agricultural yield and, in turn, improve their financial status. Fertilizer misuse costs farmers a great deal of money as well.
In today's complex environment, accurate and effective plant disease control starts with accurate diagnosis. As smart farming has expanded, plant disease diagnosis has gone digital and data-driven, enabling sophisticated decision support, astute analysis, and strategic planning. This work presents a deep learning based mathematical model for plant disease diagnosis and recognition that improves training efficiency, accuracy, and generality. Our assignment suggests a deep learning-based model that will be developed using a dataset including images of crop leaves in both good and unhealthy conditions. All algorithms' impacts are quantified using a range of metrics, such as measure, precision, and accuracy. By comparing events that were properly and erroneously predicted, accuracy is determined. The findings demonstrate that the Decision Tree algorithm achieves the other approaches in terms of presentation, with the greatest precision of 86%, and requires less time to construct the model. Thus, by employing our method, farmers may produce new crops at different times of the year, increase their profits, and avoid contaminating the land.
© 2024, IRJET
|
Impact Factor value: 8.226
Crop selection errors are common among farmers; examples include planting in the wrong season or choosing a crop that would not yield much on a given soil. A decreased yield is always the result of improper crop selection. Farmers have uncertainty over which crop to plant during a given season due to the unpredictable nature of the environment. The use of distinct fertilizers is also subject to seasonal fluctuations in the surrounding environment and the accessibility of vital components like as air, water, and soil. To get a larger yield, we need to consider not only rainfall, which is one of the key components, but also soil type and fertility characteristics. While dry land is better fit for cash crops, wetland is outstanding for wheat and sugarcane. fifteen regions of agro weather India's regions are divided based on the
|
ISO 9001:2008 Certified Journal
|
Page 829