Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 04 | Apr 2022

p-ISSN: 2395-0072

www.irjet.net

Soil Profile Based Crop Prediction System Mudra Verma1, Kaushal Vyas2, Kunal Sahjwani3 , Mahendra Patil4 1,2,3Student,

Dept. of Computer Engineering, Atharva College of Engineering, Maharashtra, India. Professor, Dept. of Computer Engineering, Atharva College of Engineering, University of Mumbai, Maharashtra, India. ---------------------------------------------------------------------***--------------------------------------------------------------------4Assistant

Abstract - In many parts of India, farmers are

current natural conditions was performed. This gives the farmer a variety of crops that can be grown. Therefore, the project develops a system by combining data from various sources, data analysis, and forecasting analysis, improving crop production and increasing the profit margins of farmers, helping them over time.

experiencing problems with crop production due to soil and climate. There is no proper guide available to help them develop the right types of plants using modern technology. Due to illiteracy, farmers may not be able to take advantage of the scientific advances made in agriculture and still cling to human practices. This makes obtaining the desired yield difficult. For example, crop failure may be due to misuse of fertilizers or unwanted rainfall patterns. In such cases, the appropriate solution would be to select crops that are suitable for current soil conditions and the expected rainfall during planting. Therefore, we are introducing a ‘Soil-Based Profile Profiling System’ based on data mining. We provide a list of crops that a farmer can cultivate based on soil input parameters (NPK and pH) and rainfall of the farmer's area. In addition, it also proposes fertilizer that can be used to improve soil quality and thus bring more crops under successful cultivation. This desktop application is designed to solve a growing problem of crop failure.

The farmer must take care of the soil in order to have a good harvest. Growers should be aware of the macronutrients and micronutrients present in the soil in order to get the maximum yield of a particular crop and know which fertilizer to use. Soil analysis is an important aspect of Cultivation. Most people do not know how to plant crops properly and in the right location. By analyzing parameters such as Sodium (N), Potassium (K), Phosphorus (P), and soil pH value, region, and rainfall, our project thus identifies the suitability of certain soil-based plants. mentioned above.

1.1 Proposed Solution

Key Words: Nitrogen, Phosphorus and Potassium, Artificial Neural Networks, pH, depth, temperature, rainfall, Feedforward backpropagation, Soil.

We proposed a desktop application that takes input as nutrients values such as N, P, K, ph and finds output as the fertility level of the soil, which crops will yield in that fertility level after adding fewer nutrients which crops can yield in that soil. Most farmers in India don’t know about their soil fertility level. So he can’t understand which crops to grow in that soil. So our system is most helpful to the farmers who don’t know about their soil fertility level and which crops they have to see in that soil.

1. INTRODUCTION Agriculture is an important activity in India, providing 118.6 million farmers with a livelihood by the 2011 census. Understanding soil conditions, knowing when and where to apply compost, considering rainfall, maintaining crop quality, and understanding how different factors work differently in different parts of the same field are some of the many problems farmers face before. as when plowing. A number of factors and statistics need to be taken into account when making important agricultural decisions that may be difficult to implement in their own right or at times. Implementing this program will provide a solution for Agriculture by monitoring the agricultural field, which can help farmers to grow their productivity on a large scale. Weather data obtained online, such as rainfall reserves and soil boundaries, provides insight into which plants should be planted in a particular area. This function introduces a program in a desktop-based application, which uses data analysis techniques to predict the most profitable yields in the current climate and soil conditions.The program will integrate data obtained from storage and climate department. Using a machine learning algorithm, predicting the most relevant plants based on

Impact Factor value: 7.529

We took readily available datasets of crops and nutrients values to implement this system. We trained this database using SVM. When the user enters an input we process the test data and compare it with the trained dataset using the SVM algorithm and check our final output.

1.2 Problem Formulation The system will integrate data obtained from storage, in the weather department and using a machine learning algorithm, the most accurate crop prediction based on current environmental conditions is made. This gives the farmer a variety of crops that can be grown. Therefore, the project improves the system by combining data from various sources, analyzing data, analyzing forecasts that can improve crop production and increase farmers' profit margins to help them over time.

ISO 9001:2008 Certified Journal

Page 3259