Python Arduino ESP32 Blynk License: MIT

Personal IoT project integrating ESP32, ESP32-CAM, and a Python OCR Flask server, connected to Blynk for control and telemetry. ESP32 detects a vehicle, triggers ESP32‐CAM to capture, ESP32‐CAM POSTs the image to Flask for OCR, results are logged to Google Sheets/Drive and sent back to ESP32 to drive the gate servo.

System Circuit Design

• ESP32 telemetry and control: Ultrasonic detection, buzzer/LED alerts, gate servo.
• ESP32‐CAM image capture: On-demand capture endpoint that forwards frames to the server.
• Python OCR server: Flask endpoint runs Tesseract OCR, uploads to Google Drive, logs to Google Sheets.
• Blynk integration: Virtual pins dashboard and notification on detection.

• Hardware: ESP32, ESP32‐CAM (AI Thinker)
• Cloud/App: Blynk (cloud + mobile app)
• Firmware: Arduino/PlatformIO (C/C++)
• Server: Python 3.x, Flask, Tesseract OCR, Google Sheets/Drive APIs

• ESP32: detects vehicle; on detection sends GET /capture to ESP32‐CAM; receives OCR via GET /receive_text?text=....
• ESP32‐CAM: handles GET /capture, takes a snapshot and POSTs raw JPEG to Flask POST /capture.
• Flask OCR server: saves image, runs OCR, overlays text, uploads to Google Drive, appends to Google Sheets, notifies ESP32.
• Blynk: dashboard via virtual pins and an optional vehicle_detected event.

• ESP32‐CAM (port 80): GET /capture
• Flask (port 5000): POST /capture, GET /uploads/latest_vehicle_annotated.jpg
• ESP32 (port 80): GET /receive_text?text=ABC123

firmware/
 esp32/
 main.ino
 esp32-cam/
 main.ino
 camera_pins.h
scripts/
 python/
 main.py
 requirements.txt
screenshots/
 demo.gif
 system-circuit-design.png

• Hardware: ESP32 dev board, ESP32‐CAM (OV2640), ultrasonic sensor, servo, buzzer/LED, USB‐TTL as needed.
• Software:
  • Arduino IDE (or PlatformIO) with ESP32 board support
  • Blynk account/app and device BLYNK_AUTH_TOKEN
  • Python 3.9+ (recommended), Tesseract OCR installed

1. Create a device in Blynk Cloud and get BLYNK_AUTH_TOKEN.
2. Virtual pins used by the firmware:
   • V0: timestamp text
   • V1: OCR plate text
   • V2: servo mode switch (0°/90°)
   • V3: manual capture trigger
3. Optional event: vehicle_detected for notifications.

• ESP32 (firmware/esp32/main.ino): set ssid, password, and BLYNK_AUTH_TOKEN.
• ESP32‐CAM (firmware/esp32-cam/main.ino): set Wi‐Fi creds and flaskServerUrl (host:port of Flask).
• Build and upload each sketch using Arduino IDE or PlatformIO.

Install Tesseract OCR (update the path in scripts/python/main.py if different). Place your Google service account credentials.json in repo root.

python -m venv .venv
.venv\Scripts\activate
pip install -r scripts\python\requirements.txt
python scripts\python\main.py

Do not commit secrets. Keep credentials.json, .env, and Arduino secrets.h out of git. Use scripts/python/example.env as a template for your local .env and create secrets.h.example files for Arduino credentials.

1. Start the Flask OCR server on a reachable host (POST /capture on port 5000).
2. Power the ESP32‐CAM; test http://<esp32-cam-ip>/capture to verify capture + POST.
3. Power the ESP32; it will trigger capture on detection and receive OCR via /receive_text.
4. Use the Blynk app to view time (V0), plate text (V1), toggle servo (V2), trigger manual capture (V3).

• Match all IPs/ports across ESP32, ESP32‐CAM, and Flask server.
• Verify Blynk token, Wi‐Fi credentials, and virtual pin mapping.
• Confirm Tesseract installation path and Google API credentials/permissions.
• If capture fails, reduce ESP32‐CAM frame_size or ensure PSRAM is available.

This project is licensed under the MIT License. See LICENSE for details.

Reno-03 and Aru (2025)