Course Content

Abstract :

Prediction of soil moisture in advance is useful to the farmers in the field of agriculture. Irrigation in agricultural lands plays a vivacious role in water and soil conservation. Future prediction of soil moisture content using real-time soil and environmental parameters may provide an efficient platform for agriculture land irrigation requirements. In this project we have used machine learning techniques such as linear regression and random forest regressor for prediction of soil moisture. These techniques were applied on datasets collected from online repositories. Here mainly we are going to predict the soil moisture content based on temperature and humidity of the placed by collecting those data in real time with the help of open weather map API .The performance of the predictor is evaluated on the basis R2 score. The comparison result shows that random forest regressor is superior providing  82 %accuracy.

 

 

Introduction :

Irrigation is a crucial practice in several agricultural cropping systems in semiarid and arid areas, and also useful water applications and management are key concerns. The efficiency and uniformity of irrigation could be maintained from the complex and diverse information based systems by considering weather, soil, water, and crop data. Sustainable agriculture, in terms of food security, rural employment, and environmentally sustainable technologies such as conservation of soil, sustainable natural resource management, biodiversity protection as well as implementation of modern agriculture practices, are crucial for holistic rural development. Irrigation water management forms a major part of precision agriculture. It involves better assessment of need and availability of soil water level for crop cultivation.

Fresh-water scarcity is one of the biggest problems faced by many nations in the world and the problem is getting adverse with time due to increase in population and inefficient usage of fresh water. Based on a United Nations report , the world population (around 7.6 billion currently) is expected to increase to 8.6 billion by 2030, 9.8 billion by 2050 and 11.2 billion by 2100 with roughly adding 83 million people yearly. Agriculture sector preoccupy a large amount of water in irrigation practices. In India, a water stressed nation, around 80% of fresh water resources are used in agriculture sector. The traditional irrigation techniques do not utilize water optimally, which raises a need for precision agriculture with smart irrigation.

Less water resources are already affecting the world and it is very important to use the water resources efficiently. Due to lack of cost-effective smart irrigation systems, developing countries are consuming more water compared with advanced countries for getting desired yield. For example, In India 2-4 times more water is using to grow a same crop which is growing in countries like Israel and Germany along with this India having worlds 2.4% of land and 18% of population. Catering food for all is difficult task so according to the nature of soil we should maintain the irrigation system in a smart way with less cost . To achieve water saving, note the supply of water from seedling to harvest so that total utilization of water for different stages of plant can updated to data sheet which contain the reading of sensor from the beginning. Water management is important for irrigation to make Sure that irrigating water is having enough minerals to grow a plant.

Growth of plant cells are affected by the scarcity soil moisture, which reduces the plant growth. Conversely excessive amount of soil moisture cause reduction in growth at certain stages of plant life. So, it is clear from these statements that proper prediction of soil moisture is one of the important parameter affecting the agricultural production. The key challenge in agriculture is to optimise the uses of natural resources in order to meet the increasing demand of crops and vegetables. Best solution for optimizing would be to integrate technological approaches for efficient use of both renewable and non-renewable resources. At the same time the system must be easy to implement and use.

So we proposed a system with the help of machine learning to predict the soil moisture content based on temperature and humidity. This system can be used to predict soil moisture based on the real time data of the place collected from the open weather map API.

 

Objective:

The main aim of this project to predict the soil moisture content based on some weather data with good accuracy with help of machine learning techniques and algorithms.

 

 

Problem Definition:

Good quality of raw materials is an important aspect on which all the agricultural industries are depended. A major part of these industries main resources are crop and vegetables. These also play an important role in economy associated with it. With the growing up of more and more agriculture based industries and with the increasing demand of food, the total yield of crops and vegetables needs to increase simultaneously. There are different factors that affect the production of good quality and quantity of crops. So it is important to understand those factors properly in order to maximize the production.

Growth of plant cells are affected by the scarcity soil moisture, which reduces the plant growth. Conversely excessive amount of soil moisture cause reduction in growth at certain stages of plant life. So, it is clear from these statements that proper prediction of soil moisture is one of the important parameter affecting the agricultural production.

 

 

 

Proposed System:

 

We proposed a system to predict soil moisture content using machine learning techniques and algorithms like linear regression and random forest regressor on real time weather data like temperature and humidity collected from open weather map API.