Embedded Project

Project Overview

This project enables object detection on the ESP32-CAM. Captured images are preprocessed, object detection is performed using a TensorFlow Lite model trained with Edge Impulse, and the results are sent to a local PHP server for display on a webpage.

Screenshots

Web Interface

Project Screenshot

ESP32-CAM Module

ESP32-CAM

Features

  • Capture Images: Use the ESP32-CAM to capture images.
  • Preprocess Images: Preprocess the captured images with TensorFlow Lite.
  • Object Detection: Perform object detection using TensorFlow Lite.
  • Post Results: Send the results to a local PHP server.
  • Display Results: Show the results on a webpage with the confidence percentage.

Technologies Used

  • ESP32-CAM: A low-cost development board with an integrated camera.
  • TensorFlow Lite: A lightweight framework for machine learning inference on microcontrollers.
  • Edge Impulse: A platform for training and deploying machine learning models on edge devices.
  • PlatformIO: Arguably the best IDE for embedded development.
  • FreeRTOS: A real-time operating system kernel for embedded devices.