Problem

  • Agriculture plays a critical role in providing food supply for growing population of the world.
  • Annual global food supply loss due to plant diseases is 40%, on average.
  • In developing countries, smallholder farmers generate more than 80% of the agricultural production. For them, the loss of crops has devastating consequences.
  • Sometimes, farmers can lose almost 100% of their crop due to plant diseases. This makes crop diseases a major threat to food security around the world.
  • Disease identification might be challenging due to the lack of the necessary lab infrastructure.

Solution

  • Timely and correct identification of a disease when it first appears is a critical step for efficient disease management.
  • An app can be developed that the farmer can have on their smartphone with the following capabilities:
    • Identify the type of plant
    • Identify if the plant has the disease
    • In case the leaves are affected by the disease, classify the disease
    • Provide treatment options (pull the information from the database with links to online sources)
  • The app can be installed on a drone to scan the entire field to improve efficiency and save time for the farmer by dramatically increasing coverage of the inspected area.
  • The app can be used to identify diseases of the plants we grow in our gardens.
Covers: theory of Plant Disease Detection
Estimated time needed to finish: 5 minutes
Questions this item addresses:
  • Why plant disease detection is important?
  • How to detect plant disease with machine learning?
0 comment
Recipe
publicShare
Star(0)

Leaf Doctor: Plant Disease Detection Using Image Classification

Contributors
Total time needed: ~2 hours
Objectives
Plant disease is one of the common problems that farmers have to deal with. In case of small farms, this can take out as much as 100% of the crop. In this capstone, an Image Classifer application is built that can detect the type of the plant and the potential disease that it might have using a deep learning architecture.
Potential Use Cases
Plant Disease Detection
Who is This For ?
INTERMEDIATEData Scientists, Machine Learning Engineers
Click on each of the following annotated items to see details.
Resources5/9
WRITEUP 1. Introduction to Plant Disease Detection using Machine Learning
  • Why plant disease detection is important?
  • How to detect plant disease with machine learning?
5 minutes
PAPER 2. Plant Disease Detection Using Machine Learning
10 minutes
REPO 3. Detection of Plant Disease using PlantVillage Dataset
10 minutes
REPO 4. Leaf Doctor - Model Repo
  • How to implement a plant disease detection model?
20 minutes
REPO 5. Leaf Doctor - Deployment Repo
  • How to package and deploy models for plant disease detection?
20 minutes
PAPER 6. Applications of image processing in agriculture: a survey
10 minutes
PAPER 7. Mask R-CNN
  • What contributions does Mask R-CNN make?
  • How well does Mask R-CNN perform?
  • What is the theory behind how Mask R-CNN works?
30 minutes
OTHER 8. Convolutional Neural Network
  • What is a convolutional neural network?
10 minutes
VIDEO 9. Leaf Doctor - Demo Video
  • How do you create a plant disease detection application?
9 minutes
Performance Resources2/2
UPLOAD_PDF 1. PlantVillage Dataset Info
  • What data we used for this project?
5 minutes
UPLOAD_PDF 2. Plant Disease Detection Model Performance
  • How did our model perform?
5 minutes

Concepts Covered

Jiri Stodulka.
Cool stuff!!! I skimmed the RECIPE fast. What do you use MASK RCNN for? Is it leaf segmentation before you apply the disease classifier?