Wildfire Detection Demo

This repository demonstrates and compares wildfire-segmentation approaches using local ONNX models.

Structure

• quartus_project/: Quartus project folder including vhdl-model.
• python_uart/: Python script to run inference on FPGA using PC.
• model_training/: Training scripts, ONNX export scripts, and utility scripts.
• models/: Exported model files (.onnx, .tflite).
• comparison/: Benchmark and video comparison script, plus generated output videos.
• video/: Input video used for inference.
• dataset/: Training, validation, and test data.

Running Model on FPGA

1. Programm FPGA

The Quartus project is set for an ACX3000 board with an A3CY100BM16AE7S. Compile quartus project using Quartus 25.3 and programm FPGA.

2. Run uart_cnn_visualizer.py

Adapt line 22 of the uart_cnn_visualizer.py script according to the COM port of your board. Then run the script:

python .\uart_cnn_visualizer.py

The script provides further instructions on how to use it. It is possible to either run inference on a video stream (provided by a webcam connected to the PC) or on example images contained in the img/ -folder.

Running Model Comparison

1. Model Comparison and Video Inference

(First unpack the models from models/models.zip)

Runs one or multiple ONNX segmentation models on video/demo_video.mp4, overlays masks, and writes output videos.

Run all models:

python comparison/comparison.py --model all --threads 1

Run a single model:

python comparison/comparison.py --model oneai --threads 1
python comparison/comparison.py --model unet --threads 1
python comparison/comparison.py --model deeplab --threads 1

Optional thresholds:

python comparison/comparison.py --model all --thr_oneai 0.5 --thr_unet 0.5 --thr_deeplab 0.5

Output:

• comparison/output_CustomCNN.mp4
• comparison/output_UNet.mp4
• comparison/output_DeepLabV3+.mp4

2. Theoretical Operations (FLOPs / Parameters)

Calculates theoretical Conv FLOPs and parameter counts from ONNX graph metadata.

python comparison/comparison.py --ops --model all

3. Training and ONNX Export

UNet training (wildfire_demo.py)

python model_training/wildfire_demo.py --epochs 100 --export_onnx

DeepLabV3+ MobileNet training (wildfire_demo_deeplab.py)

python model_training/wildfiren_demo_deeplab.py --epochs 300 --export_onnx

4. Count ONNX Parameters

Counts total parameters from ONNX initializers.

python model_training/count_onnx_params.py

Requirements

• numpy
• opencv-python
• onnxruntime
• onnxscript
• onnx
• torch
• torchvision
• pillow

Install:

pip install numpy opencv-python onnxruntime onnx onnxscript torch torchvision pillow