Agriculture prediction analysis system
Abstract : Agriculture is the backbone of Indian economy. Due to global warming and climate change traditional farming in the regular months have been distorted and crops have been ruined is the most common phrase seen today. This not only gives economic losses but also the main reason for farmer sucide. Now agriculture needs support, time has come for technology to take over change. For a crop to grow, favourable soil conditions, ambient rainfall and temperature is necessary. So as now due to climate change temperature and rainfall cannot be well defined , example rains in December and January or irregular temperatures have made it difficult for farmers and common man to predict months of plantation and yield of the crop due to irregularities. So we have formulated an analysis by prediction of a favourable crop based on temperature and current rainfall with soil conditions. Data science have proved that data which we have plays a vital role in predictions and iot based applications. Data science in agriculture is a growing field and has a wide scope in future.
? Agriculture is the main occupation in India and economy of the country is entirely depended on it for rural based existence.
? The primary goal of this exploration workis to give a technique with the goal that it can perform illustrative examination on crop yield creation in a compelling way.
? It is difficult to achieve smart farming in developing country because many of the farmers are illiterate and unaware of the technology.
? It implies organization of information whatever, our framework should work with same proficiency.
? It finds the near optimum minimum of error and increases the accuracy of the prediction.
? The impact of climate change are most evident in crop productivity because this parameter represents the component of greatest concern to producers, as well as consumers.
? There is no aspect of crop culture that is immune to impact of weather.
? This issue of the ranchers has been settled through accuracy farming.
? This technique is described by a dirt database gathered from the ranch, crop gave by farming specialists, accomplishment of parameters, for example, soil by soil testing lab datasets.
? Crop yield prediction has been a challenging issue for farmers since many years.
• The proposed system is not trained on a larger set of features like (climate and soil) along with time series image data to tune the trained model for accuracy.
• It is a near investigation that tells the precision of preparing proposed model and blunder rate.
• The proposed model can give the real expense of assessed crop yield and it is name as LOW, MID, and HIGH.
• Multilayer perceptron shows the most elevated exactness among the proposed calculations.
• This research work a hybrid MLRANN model was proposed for effective crop yield prediction.
? The prediction accuracy of the hybrid model is compared with the models ANN, MLR, Support Vector Regression (SVR),k- Nearest Neighbor (KNN), and Random Forest (RF) using performance metrics.
? This results in providing the farmers with efficient information needed to obtain high yield and thus maximize profits.
? Machine Learning algorithms were used to predict the crops.
? Convolutional Neural Networks (CNNs) are used in this analysis to create a crop yield prediction model based on the Normalized Difference Vegetation Index (NDVI) and RGB data acquired from Unmanned Aerial Vehicles (UAVs).
? Using supervised and unsupervised learning algorithms, such as BPN (Back Propagation Network) and Kohonen Self Organizing Map (Kohonen's SOM) are used for prediction of soil quality.
|