DEPARTMENT OF COMPUTER ENGINEERING

Agricultural Recommendation System using Deep Learning

Avishkar Dalvi 20202002

Riddhi Narkar 19102003
Radha Rakshe 19102067
Aarya Totey 19102070

Under the Guidance of: Prof. Deepak Khachane

Problem statement

To implement a Machine Learning system that gives recommendations for Crop-Fertilizers pair based on

the soil and weather conditions, also helps in identifying plant disease to provide a cure for the same

using Deep Learning technology.

Agricultural Recommendation System using DL

Project Features

● EXISTING FEATURES

Crop Based Recommendation system – Recommends the favorable crops as per the soil details.

● NEWLY PROPOSED FEATURES:-

Fertilizer Recommendation system – The user provides the details about the soil and system will give
recommendations and solutions to improve the fertility of the soil.

Crop disease classification – The user needs to upload photos of the crop affected by the disease and
the proposed system will classify the disease and will give recommendation to the farmer about how to
cure the crop.

Agricultural Recommendation System using DL

Objective

● To provide a single place to access help for crop related problems, the causes, and how to cure them,
thus providing a solution based approach

● To use recent technologies like Machine Learning’s powerful algorithms to help farmers

● To make these services available over the internet to anyone who wants to access them

● To try to make the UI as simple as possible so that a naïve user who doesn’t possess much
knowledge about technology can easily navigate

Agricultural Recommendation System using DL

Scope

● This project has the potential to assist a farming enthusiast or a occupational farmer to make smarter
decisions on his harvest strategy

● It can provide necessary guidance related to cultivation and fertilizer ideas and curing crop diseases

● Since it is easily accessible for anyone with an Internet connection, it can gather more reach

Agricultural Recommendation System using DL

Technology Stack
● Python 3.8 or higher

● Numpy

● Pandas

● Scikit

● Matplotlib

● PyTorch

● Flask

● HTML

● CSS

Agricultural Recommendation System using DL

References
Title of the research paper Year Authors Abstract
published

Plant Leaf Disease 2020 Melike Sardogan, Adem The early detection of diseases is important in agriculture for an efficient crop yield. The bacterial
Detection and Classification Tuncer, Yunus Ozen spot, late blight, septoria leaf spot and yellow curved leaf diseases affect the crop quality of
based on CNN with LVQ tomatoes. Automatic methods for classification of plant diseases also help taking action after
algorithm Department of Computer detecting the symptoms of leaf diseases. This paper presents a Convolutional Neural Network
Engineering, Yalova University (CNN) model and Learning Vector Quantization (LVQ) algorithm based method for tomato leaf
IEEE Explore disease detection and classification. The dataset contains 500 images of tomato leaves with four
symptoms of diseases. We have modeled a CNN for automatic feature extraction and
classification. Color information is actively used for plant leaf disease researches. In our model,
the filters are applied to three channels based on RGB components. The LVQ has been fed with
the output feature vector of convolution part for training the network. The experimental results
validate that the proposed method effectively recognizes four different types of tomato leaf
diseases.

Crop Recommendation 2020 Pradeepa Bandara, Recommendation system through integrated models of collecting environmental factors
System Thilini Weerasooriya, using Arduino microcontrollers, Machine learning techniques such as Naïve Bayes
Ruchirawya T.H., (Multinomial) and Support Vector Machine (SVM), Unsupervised machine learning
International Journal of W.J.M. Nanayakkara, algorithm such as K-Means Clustering and also Natural Language Processing (Sentiment
Computer Applications Dimantha M.A.C, Analysis) concerned with the Artificial Intelligence to recommend a crop for the selected
(0975 – 8887) Pabasara M.G.P. land with site-specific parameters with high accuracy and efficiency. It has been a major
problem to identify what to grow, any man has adequate space in the owner’s land. Not
Volume 175– No. 22, Sri Lanka Institute of only domestic lands but also for farming lands. Why it has become a problem is that
October 2020 Information Technology, Sri environmental factors such as temperature, water levels, and soil conditions are uncertain
Lanka as they change from time to time. Due to these problems, this solution of crop
recommendation system predicts the user, what crop type would be the most suitable for
the selected area by collecting the environmental factors for plant growth and processing
them with the trained sub-models of the main of the system.

Agricultural Recommendation System using DL

References
Title of the research paper Year Authors Abstract
published

Machine Learning based 2021 Sayed Mazhar Ali, In Pakistan, the most part of the land is used for agriculture cultivation to meet the desires of
Crop Recommendation Bhagwan Das, nearby people and export want as properly. Crop cultivation anywhere in the world depends on
System for Local Farmers of Dileep Kumar the climate so called seasons and soil properties, however, the enhancing the production of
Pakistan crops depend on various factors like mainly on temperature. In order to address the issue of
Department of Electronic increasing crop production for Pakistan, a crop recommendation system is proposed in this work.
A Research Gate Preprint Engineering, Quaid In this work, idea of ideal harvest prior to planting it, it would be of extraordinary assistance to the
Online Publication University of Engineering, farmers and others required to settle on fitting choices on upgrading the creation of yields for
Science and Technology, neighborhood utilization needs and may prompt the capacity and expanded fare choice for
Nawabshah, Sindh, Pakistan. business. Our framework utilized Machine Learning procedures with the end goal that it proposes
the appropriate corps dependent on the temperature. This framework subsequently diminishes
the monetary misfortunes looked by the farmers brought about by establishing the ominous
harvests and furthermore it gives the information on the occasional characterization of yields
what harvest is reasonable for which season. It is concluded that proposed algorithm has an
average accuracy of 90% on the given dataset. The achieved accuracy is more in comparison to
existing work.

Machine Learning Based 2021 Dhruv Piyush Parikh, The three most basic amenities required for the survival of a human being are food,
Crop Recommendation Jugal Jain, Tanishq Gupta, shelter and clothing. In today’s tech-savvy generation, the latter two have witnessed a
System Rishit Hemant Dabhade huge scientific boost. Unfortunately, even today, agriculture is considered as more of a
man-power oriented field. Most of the farmers are untutored and have little to no scientific
International Journal of Vellore Institute of Technology, knowledge of farming. So, they have to rely on the hit and trial method to learn from
Advanced Research in VIT Chennai, India experience which leads to wastage of time and resources. Our system focuses on
Science, Communication building a predictive model to recommend the most suitable crops to grow in a particular
and Technology farm based on various parameters. This can be helpful for the farmers to be more
productive and competent without wasting any resources by farming the most competent
crops.

Agricultural Recommendation System using DL

References

Title of the research paper Year Authors Abstract

published

Using Deep Learning for 2019 Sharada P. Mohanty, David P. Crop diseases are a major threat to food security, but their rapid identification
Image-Based Plant Hughes, and Marcel Salathé remains difficult in many parts of the world due to the lack of the necessary
Disease infrastructure. The combination of increasing global smartphone penetration and recent
Detection Digital Epidemiology Lab, advances in computer vision made possible by deep learning has paved the way for
EPFL, Geneva, Switzerland, smartphone-assisted disease diagnosis. Using a public dataset of 54,306 images of
Frontiers in Plant Science School of Life Sciences, EPFL, diseased and healthy plant leaves collected under controlled conditions, we train a deep
Online Publication Lausanne, Switzerland. convolutional neural network to identify 14 crop species and 26 diseases (or absence
thereof). The trained model achieves an accuracy of 99.35% on a held-out test set,
demonstrating the feasibility of this approach. Overall, the approach of training deep
learning models on increasingly large and publicly available image datasets presents
a clear path toward smartphone-assisted crop disease diagnosis on a massive global
scale.

Agricultural Recommendation System using DL

Gantt Chart

Agricultural Recommendation System using DL