Skip to content

Latest commit

 

History

History
70 lines (39 loc) · 2.04 KB

README.md

File metadata and controls

70 lines (39 loc) · 2.04 KB
Raw

SmartYields 🌱

A Crop Recommendation and Crop Disease Detection System using Machine Learning, Big data Analysis

image69

Crop Recommendation based on factors:

  • Temperature
  • Humidity
  • pH
  • Rainfall
  • N, P, K - Nitrogen, Phosphorous, Potassium content of soil

Screenshots

Disease Detection

Diseased crop image

image36

Result

image73

Healthy crop image

image132

Result

image5

Crop Recommendation

Inputs

image98

Result

image135

Models Used:

  1. Crop Recommendation: Random Forests (98.51% accuracy)
  2. Disease Detection: ResNet (98% accuracy)

Modeling

Disease Detection

image61

Crop Recommendation

image113

Tools/Languages

  • Jupyter Notebook: For data cleaning, analysis, profiling, modeling, visualizing, recording different results.
  • Tensorflow: For image recognition
  • Libraries/Modules Used: Pandas, Seaborn, Matplotlib, Sklearn, Pickle.

Web app:

  • Flask: Framework to use the predict function from pickle file and build middleware and frontend.
  • Bootstrap: Framework used for presentation and styling the frontend.