How ajaybhashaveni/robotic-arm-sprinkler-for-pesticide-spraying works
This repository supports a paddy-leaf disease model workflow: prepare the three-class image dataset, train a classifier, then quantize and package it for Arduino use. The README specifically documents dataset extraction through the notebook; the sampled Python source documents model training and conversion. The Arduino sketch is present, but its runtime behavior and connection to the generated header were not sampled, so those links are left unspecified.
Main entry points: convert_to_tflite.py TFLite converter · train_model.py Model trainer · train_model.ipynb Training notebook · model_data.h Model C header
Components 9
Dataset preparation
- Training notebook
train_model.ipynb - Class image dataset
Model training
- Model trainer
train_model.py - Keras model
Embedded deployment
- TFLite converter
convert_to_tflite.py - Quantized model
- Model C header
model_data.h
Robot runtime
- Plant robot sketch
plant_robot.ino
Other components
- Dataset provider
Connections 8
- Dataset provider → Training notebook: provides ZIP
- Training notebook → Class image dataset: extracts classes
- Model trainer → Class image dataset: loads images
- Model trainer → Keras model: saves model
- TFLite converter → Keras model: loads model
- TFLite converter → Class image dataset: samples images
- TFLite converter → Quantized model: writes model
- TFLite converter → Model C header: writes array