crop yield prediction and fertilizer recommendation

Crop Yield Prediction And Fertilizer Recommendation
ABSTRACT
Agriculture is first and foremost factor which is important for survival. The Indian economy,
dominatingly, depends upon farming along with agro-industry things. The soil ingredients
(like Nitrogen, Phosphorous, Potassium), crop rotation, soil clamminess, atmospheric and
surface temperature, precipitation, etc, play an efficient role in cultivation. Deep Learning
has emerged as a potential tool for crop yield prediction. Recurrent Neural Network that consists
of multiple hidden layers between the input and output layers. Recurrent Neural Networks (Many
to One RNNs). Using Recurrent Neural Networks many inputs are given in input layers and
generate output in hidden layer to produce single output. The performance parameter accuracy
and scalability to be improved. The proposed model is enhanced by applying deep learning
techniques and along with the prediction of crop, a clear information is achieved regarding
the amounts of soil ingredients needed with their expenses separately. It provides a better
accuracy than the existing model. It analyzes the given data and help the farmers in predicting a
crop which in return help in gaining profits.
EXISTING SYSTEM
• In existing system, a full 3-D convolutional neural network is constructed to maximally

explore deep spatial–spectral features from multispectral images.
• Image based prediction may provide inaccuracy of prediction which leads several issues
to the farmers.
• Hence, there is a need for new technologies and methods to enhance the existing system
in order to get faster and better results.
Disadvantages of existing system
 No prompt and accuracy

 Over fitting may be occurred
Problem Definition
 Previous algorithms can build empirical predictive models using a fixed pattern of
features extracted from remote sensing images without abundant parameters. However,
the performance is still limited because handcrafted features mainly leverage human
ingenuity and prior knowledge, which are not robust for all cases.
 Furthermore, such algorithms are computationally intensive and become intractable on

large-scale datasets. All such difficulties make crop yield prediction challenging and lead
to unsatisfactory performance.
 Less cultivation.
 Less profit.
 No proper updation by the farmer in scientific way.

PROPOSED SYSTEM
• Predictive analytics is about analyzing current and historical data to forecast the
probability of future events, outcomes, or values in the context of crop predictions.
• Predictive analytics requires numerous statistical techniques, such as data mining

(identification of patterns in data) and deep learning such as recurrent neural network.
• We proposed a deep learning model that would predict and suggest a crop to the user
based on the features of the soil along with the climatic conditions of the region.
• The dataset attributes are nitrogen percentage, phosphorus percentage, potassium

percentage, rainfall, humidity, temperature and PH. The dataset is first preprocessed, then
trained and then tested.
Advantages of Proposed System
 Recurrent Neural Network technique provides more accuracy of prediction.

 Comparison of multiple algorithms may provide a knowledge about best accuracy..
SYSTEM SPECIFICATION
SOFTWARE REQUIREMENTS
The below Software Specifications were used in both Server and Client machines when
developing.
SERVER
Operating System : Windows 10/11

Technology Used : Python
Web Server : Django
Browser : Google Chrome
IDE : Jetbrains PyCharm
CLIENT
Operating System : Windows 10/11

Browser : Chrome
HARDWARE REQUIREMENTS
The below Hardware Specifications were used in both Server and Client machines when
developing.
Processor : Intel(R) Core(TM) i3

Processor Speed : 3.06 GHz
RAM : 2 GB
Hard Disk Drive : 250 GB
CD-ROM Drive : Sony
Monitor : “17” inches
Keyboard : TVS Gold
Mouse : Logitech
MODULE DESCRIPTION
Module – I: Gathering and Analyzing Data :

The large data set has been collected from various sources on crops and their climatic and
soil requirement and analyzed.
Module – II: Data Pre-Processing

In which data are splitted and remove the wrangling data. Then they will be passed to
process of feature extraction in which text data converted into numerical format of data to
understand the machine before executing.
Module – III: Data Clustering

Recurrent neural network model are trained and tested to produced an high defined
accuracy in crop selection.
Module – IV: Predicting the Crop

After testing, the next step is predicting the crop. The user knows the exact crop to be
deployed for the current geographical status.

crop yield prediction and fertilizer recommendation

Uploaded by

Copyright:

Available Formats

crop yield prediction and fertilizer recommendation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

crop yield prediction and fertilizer recommendation

Uploaded by

Copyright:

Available Formats

Crop Yield Prediction And Fertilizer Recommendation

• In existing system, a full 3-D convolutional neural network is constructed to maximally

Disadvantages of existing system

 No prompt and accuracy

 Furthermore, such algorithms are computationally intensive and become intractable on

 No proper updation by the farmer in scientific way.

• Predictive analytics requires numerous statistical techniques, such as data mining

• The dataset attributes are nitrogen percentage, phosphorus percentage, potassium

Advantages of Proposed System

 Recurrent Neural Network technique provides more accuracy of prediction.

Operating System : Windows 10/11

Operating System : Windows 10/11

Processor : Intel(R) Core(TM) i3

Module – I: Gathering and Analyzing Data :

Module – II: Data Pre-Processing

Module – III: Data Clustering

Module – IV: Predicting the Crop

You might also like