CV / License Plate

Tiny-YOLOv4 License Plates

reComputer R Series (CM5 + Hailo-10H)

Single-class license-plate detection on reComputer R Series with Hailo-8 or Hailo-10H.

55 downloads

Size

8.21 MB

Memory

4GB+

Precision

Hailo HEF / HailoRT 4.23.x (Hailo-8) / 5.1.1 (Hailo-10H)

Choose the device you're using, the set up guide and documentation will update accordingly.

Platform

Inference Engine

Getting Started

Deploy

sudo docker run --rm \
  --name cm5-hailo10h-plate-detector \
  --privileged \
  --net=host \
  -e PYTHONUNBUFFERED=1 \
  --device /dev/hailo0:/dev/hailo0 \
  -v /usr/lib/libhailort.so.5.1.1:/usr/lib/libhailort.so.5.1.1:ro \
  -v /usr/lib/libhailort.so:/usr/lib/libhailort.so:ro \
  ghcr.io/seeed-projects/recomputer-hailo10h-cv/tiny_yolov4_license_plates:latest \
  python web_detection.py --model_path model/tiny_yolov4_license_plates.hef --video_path video/test.mp4

REST API

Use the REST API to run inference. Copy the commands below.

Curl

curl -X POST "http://<Board_IP>:8000/api/models/tiny_yolov4_license_plates/predict" \
  -F "file=@traffic.jpg"

Python

import requests

response = requests.post(
    "http://<Board_IP>:8000/api/models/tiny_yolov4_license_plates/predict",
    files={"file": open("traffic.jpg", "rb")},
    timeout=30,
)
print(response.json())

Model Details

reComputer R Series (CM5 + Hailo-10H)

Tiny-YOLOv4 License Plates on reComputer CM5 + Hailo-10H

Tiny-YOLOv4 detects license plates (single class) on Hailo-10H through HailoRT 5.1.1. The HEF exposes 2 raw YOLO heads; anchor decode and NMS run on the CPU.

Model Info

Property	Value
Task	License plate detection (single class)
Input	416x416x3
Output	13x13x18 and 26x26x18 raw YOLO heads
HEF	Hailo Model Zoo v5.4.0, Hailo-10H (5.1 MB)

Hardware and Host Setup

Item	Value
Board	reComputer R Series with Raspberry Pi CM5
Accelerator	Hailo-10H over PCIe, exposed as `/dev/hailo0`
Host driver	`hailo-h10-all` apt package
Runtime	HailoRT 5.1.1
Python in container	3.13, aarch64

bash

sudo apt update
sudo apt install hailo-h10-all -y
sudo reboot

# After reboot
hailortcli fw-control identify
ls /dev/hailo0

Run With Demo Video

bash

sudo docker run --rm \
  --name cm5-hailo10h-plate-detector \
  --privileged \
  --net=host \
  -e PYTHONUNBUFFERED=1 \
  --device /dev/hailo0:/dev/hailo0 \
  -v /usr/lib/libhailort.so.5.1.1:/usr/lib/libhailort.so.5.1.1:ro \
  -v /usr/lib/libhailort.so:/usr/lib/libhailort.so:ro \
  ghcr.io/seeed-projects/recomputer-hailo10h-cv/tiny_yolov4_license_plates:latest \
  python web_detection.py --model_path model/tiny_yolov4_license_plates.hef --video_path video/test.mp4

REST API

bash

curl -X POST "http://<Board_IP>:8000/api/models/tiny_yolov4_license_plates/predict"   -F "file=@plate.jpg"

Development Notes

Source module: src/hailo10h_tiny_yolov4_license_plates/
Dockerfile: docker/hailo10h/tiny_yolov4_license_plates.dockerfile
Container: ghcr.io/seeed-projects/recomputer-hailo10h-cv/tiny_yolov4_license_plates:latest
Executor is the HailoRT 5.1.1 multi-output create_infer_model wrapper.
HEF: Hailo Model Zoo v5.4.0 hailo10h build.

Inputs and Outputs

Input: 416x416 RGB traffic frame. Output: license-plate confidence and xyxy box after raw-head decoding and CPU NMS.