Skip to content

README.md

Latest commit

 

History

History
20 lines (14 loc) · 782 Bytes

README.md

File metadata and controls

20 lines (14 loc) · 782 Bytes
Raw

Soil-Moisture-Analysis-Using-CNN

Classifies soil images as dry or wet using a CNN model with 98.6% accuracy. Compares predictions with sensor data to support smart irrigation. Includes a Streamlit dashboard for visualizing model vs sensor output using gauges and line graphs. Combines deep learning and IoT for precision farming.

🔍 Features

  • Classifies soil as dry, moist, or wet based on image data.
  • Uses a custom-trained CNN built with TensorFlow/Keras.
  • Preprocessing pipeline for cleaning and resizing images.
  • Includes model training, evaluation, and inference scripts.

🧪 How to Run

  1. Clone the repo: 
    git clone https://github.com/tzprograms/Soil-Moisture-Analysis-Using-CNN.git
cd Soil-Moisture-Analysis-Using-CNN