pp-nsfw_Inspector

pp-nsfw_Inspector(中文版)

pp-nsfw_Inspector is an image content moderation pipeline running on the Axera NPU. It combines OCR, NSFW detection, QR code scanning, and keyword rule matching to classify images as PASS / REVIEW / REJECT.

---

Pipeline Overview

flowchart TD
    A[Input Image] --> B[Preprocess Layer<br/>Scene Classification + Image Processing + Long Image Slicing]
    B --> C{Process by Slice}

    C --> D[Image Branch<br/>NSFW + QR Code]
    C --> E{OCR Routing}

    E -->|SCREENSHOT| F[PP-OCRv5]
    E -->|DOCUMENT / POSTER / UNKNOWN| G[PP-DocLayout-S]
    G --> H[Text Region OCR]
    G --> I[Figure Region Extraction]
    I --> J[Figure Region NSFW]

    F --> K[OCR Result<br/>blocks + avg_score + text_state]
    H --> K
    D --> L[Image Signals<br/>nsfw / qr]
    J --> L

    K --> M[Understanding Layer<br/>Text Normalization + Strong/Weak Rules]
    M --> N[Text Signals<br/>rule_hits]
    L --> O[Decision Layer]
    K --> O
    N --> O

    O --> P{Final Action}
    P -->|PASS| Q[Release]
    P -->|REVIEW| R[Review]
    P -->|REJECT| S[Reject]

---

Supported Tasks

---

Model Details

All models are exported in w8a16 quantization for Axera NPU as .axmodel format. The following data is measured with ax_run_model -r 100 -w 10 (single-model benchmark, 100 iterations, 10 warmup).

• NPU1: Single core, OCR models via Pulsar2 5.2, PP-DocLayout-S / NSFW via Pulsar2 6.0
• NPU3: Triple core, Pulsar2 6.0 (CMM size differs due to 3-core fusion strategy)

Model conversion tools: Pulsar2-docs (ver 5.2+). Engine version: 2.10.1s.

Note: Toolchain version differences may affect operator optimization, resulting in minor performance variations.

---

Support Platform

• AX650N / AX8850
  • M4N-Dock (爱芯派Pro)
  • AI Pyramid
  • M.2 Accelerator Card

---

How to Use

Python Environment

pyaxengine

# pyaxengine
wget https://github.com/AXERA-TECH/pyaxengine/releases/download/0.1.3.rc3/axengine-0.1.3-py3-none-any.whl
pip install axengine-0.1.3-py3-none-any.whl

# Other dependencies
pip install -r requirements.txt

Note: pyzbar requires the system zbar shared library.

Local Test (CLI)

python test.py

Iterates over images in images/ and prints the decision for each:

action:     REJECT
risk_level: high
labels:     ['nsfw']
score:      0.987
evidence:   [{'source': 'nsfw_model', 'score': 0.987}]

Web Demo

python app.py

Open http://127.0.0.1:5000/ in a browser. The web page supports:

• Viewing a default sample image result
• Uploading local images for moderation
• Displaying decision, risk level, labels, evidence, slice details, OCR status, normalized text, and rule hits

Note: For safety, only partial keyword anomaly detection is enabled in this public release. For the complete illegal keyword list, please file an issue or contact us through official channels.

Run in background:

setsid python app.py > web.log 2>&1 < /dev/null &
# Stop: pkill -f 'python app.py'
# Logs: tail -f web.log

---

Decision Semantics

The decision layer uses a tiered OR logic:

Strong Signals → REJECT

• Strong keyword rule hit
• QR code blacklist domain
• NSFW score at or above scene-specific reject threshold

Weak Signals → REVIEW

• Weak keyword rule hit
• QR code unknown domain
• NSFW score at or above review threshold
• OCR low quality (avg_score < 0.65 and blocks < 3)
• OCR missing expected text / missing uncertain

NSFW Thresholds by Scene

---

Output Fields

---

Limitations

• Not a production service (no auth, no access control)
• REVIEW strategy is conservative (cold-start phase)
• Rules and thresholds are in early tuning stage
• Designed for development, integration testing, and demonstrations

---

Other

• PP-OCRv5 models: AXERA-TECH/PPOCR_v5
• PP-DocLayout-S model: PP-DocLayout
• nsfw model: Falconsai/nsfw_image_detection_26
• Test Images: All images under images are website screenshots used for internal regression testing.