security-scan

Security X-ray Scan Knife Detection

Execution model

Run one phase at a time. After each phase, verify the expected result before continuing.

• If a phase succeeds → print [OK] and move to the next phase.
• If a phase fails → print [STOP], consult the failure decision tree, and ask the user before retrying.

Phase 1 — Verify prerequisites

Hardware required:

• 1-2x Raspberry Pi 4B
• reComputer J1010 (or J1020, J2011, J2012, AGX Xavier)
• HDMI display, mouse, keyboard
• All devices on the same network

# On Raspberry Pi: check Python version (need 3.9.2)
python3 --version
# On reComputer: check JetPack
cat /etc/nv_tegra_release

Expected: Python 3.9.2 on RPi; JetPack 4.6.1 on reComputer.

Phase 2 — Set up Raspberry Pi environment

# Update system
sudo apt-get update
sudo apt-get upgrade

# Install dependencies
sudo apt-get install python3-pip libjpeg-dev libopenblas-dev libopenmpi-dev libomp-dev

# Install setuptools and Cython
sudo -H pip3 install setuptools==58.3.0
sudo -H pip3 install Cython

# Install gdown for Google Drive downloads
sudo -H pip3 install gdown

# Install PyTorch 1.11.0 (Buster OS, aarch64)
gdown https://drive.google.com/uc?id=1gAxP9q94pMeHQ1XOvLHqjEcmgyxjlY_R
sudo -H pip3 install torch-1.11.0a0+gitbc2c6ed-cp39-cp39-linux_aarch64.whl
rm torch-1.11.0a0+gitbc2c6ed-cp39-cp39-linux_aarch64.whl

Verify PyTorch:

import torch as tr
print(tr.__version__)

Install remaining dependencies:

# Tritonclient
pip3 install tritonclient[all]

# TorchVision 0.12.0
gdown https://drive.google.com/uc?id=1oDsJEHoVNEXe53S9f1zEzx9UZCFWbExh
sudo -H pip3 install torchvision-0.12.0a0+9b5a3fe-cp39-cp39-linux_aarch64.whl
rm torchvision-0.12.0a0+9b5a3fe-cp39-cp39-linux_aarch64.whl

# OpenCV
pip3 install opencv-python

Expected: All packages install without errors.

Phase 3 — Set up reComputer J1010 (Triton Server)

Ensure JetPack 4.6.1 is installed on the reComputer.

# Create model repository and download ONNX model
mkdir -p ~/server/docs/examples/model_repository/opi/1
# Download model.onnx from: https://drive.google.com/file/d/1RcHK_gthCXHsJLeDOUQ6c3r0RlAUgRfV/view
# Place model.onnx into ~/server/docs/examples/model_repository/opi/1/

# (Optional) Clone general Triton server examples
git clone https://github.com/triton-inference-server/server
cd ~/server/docs/examples
sh fetch_models.sh