Detector24
Back to Model Catalogue
Gambling Detector
ImageRestricted Content

Gambling Detector

Detect gambling content in images: casinos, slot machines, poker, and betting. Enforce advertising regulations and protect underage users automatically.

Accuracy
93.8%
Avg. Speed
5.0s
Per Request
$0.0060
API Name
gambling-detection

Bynn Gambling Detector

The Bynn Gambling Detector identifies and locates gambling-related content in images, including casino equipment, playing cards, slot machines, and gambling activities. This model is essential for enforcing gambling advertising restrictions and protecting minors.

The Challenge

Online gambling advertising is heavily regulated to prevent problem gambling and protect vulnerable populations. Yet gambling content proliferates across social media, often targeting young audiences through influencer partnerships, streaming integrations, and lifestyle content that normalizes betting as entertainment.

Regulations vary dramatically by jurisdiction—outright bans in some regions, licensed advertising in others, complex restrictions everywhere. Platforms must identify gambling content accurately to enforce region-specific policies, prevent unlicensed gambling promotion, and meet their duty of care to users at risk of gambling addiction. The visual diversity of gambling content—from casino imagery to sports betting apps to poker streams—requires comprehensive detection.

For physical environments, gambling detection enables compliance and policy enforcement. CCTV systems can detect unauthorized gambling in workplaces, schools, or prohibited areas. Licensed casinos can monitor for policy violations. Facilities can detect improvised gambling setups in areas where gambling is prohibited. The same detection capabilities support both digital platform compliance and physical space monitoring.

Model Overview

When provided with an image, the detector analyzes the visual content to identify gambling-related items and activities. The model uses advanced object detection with instance segmentation, providing both the location of detected items and pixel-level masks that precisely outline each detection.

Achieving 93.8% accuracy, the model can detect a wide variety of gambling-related content including slot machines, poker tables, casino chips, playing cards, roulette wheels, and people engaged in gambling activities.

How It Works

The model employs open-vocabulary object detection technology, allowing it to recognize gambling-related items based on learned visual patterns. For each detected item, the system returns:

  • Bounding box coordinates: The rectangular region containing the detected object
  • Confidence score: A probability value (0.0-1.0) indicating detection certainty
  • Class name: The specific type of gambling-related item detected
  • Segmentation mask: A pixel-level PNG mask outlining the exact shape of the detected object

Response Structure

The API returns a structured response containing:

  • num_detections: Total count of gambling-related items detected in the image
  • detections: Array of detection objects, each containing:
    • bbox: Bounding box coordinates {x1, y1, x2, y2}
    • score: Detection confidence (0.0-1.0)
    • class_name: Type of gambling-related item
    • mask_png: Base64-encoded PNG segmentation mask
  • image_size: Original image dimensions {width, height}

Detected Classes

The model can identify the following gambling-related items:

Casino Equipment

  • Slot machine
  • Roulette wheel
  • Poker table, casino

Gaming Items

  • Playing cards, poker
  • Dice
  • Poker chips, casino chips
  • Lottery ticket, bingo

People & Activities

  • Card dealer
  • Gambling, person gambling
  • Person playing cards, card game
  • Person at slot machine

Performance Metrics

Metric Value
Detection Accuracy 93.8%
Average Response Time 5000ms
Max File Size 20MB
Supported Formats GIF, JPEG, JPG, PNG, WebP

Use Cases

  • Advertising Compliance: Enforce gambling advertising restrictions across digital platforms
  • Age-Restricted Content: Protect minors from exposure to gambling content
  • Social Media Moderation: Detect gambling promotion that violates community guidelines
  • Brand Safety: Prevent brand advertisements from appearing alongside gambling content
  • Responsible Gaming: Support responsible gambling initiatives by monitoring promotional content
  • Regulatory Compliance: Assist in meeting regional gambling advertising regulations

Known Limitations

Important Considerations:

  • Recreational vs. Gambling: Family board games with dice or cards may be detected (e.g., Monopoly, standard card games)
  • Artistic Content: Artistic or historical depictions of gambling may be flagged
  • Context Limitations: The model cannot determine if gambling is legal or occurs in a licensed venue
  • Virtual Gambling: Screenshots of online gambling may have lower detection rates than physical equipment
  • Partial Visibility: Heavily occluded items may have lower detection confidence

Disclaimers

This model provides probability scores, not definitive identification of illegal gambling.

  • Screening Tool: Use as part of a broader content moderation strategy, not as the sole decision factor
  • Human Review: High-stakes decisions should involve human review for context assessment
  • Legal Context: Gambling regulations vary significantly by jurisdiction
  • False Positives: Some recreational gaming items may trigger detections; adjust thresholds accordingly

Best Practice: Combine detection results with human review and regional legal expertise for optimal content moderation outcomes.

API Reference

Version
2601
Jan 3, 2026
Avg. Processing
5.0s
Per Request
$0.006
Required Plan
trial

Input Parameters

Open-vocabulary object detection with segmentation. Detects objects using text prompts.

image_urlstringRequired

URL of the image for object detection

Example:
https://example.com/image.jpg

Response Fields

Object detection results with bounding boxes and segmentation masks

num_detectionsinteger

Number of objects detected

Example:
3
detectionsarray

Array of detected objects

Array Item Properties:
bboxobject

Bounding box {x1, y1, x2, y2}

{"x1":100,"y1":150,"x2":300,"y2":400}
scorefloat

Detection confidence (0.0-1.0)

0.95
class_idinteger

Class index

0
class_namestring

Detected object class name

person
mask_pngstring

Base64-encoded PNG segmentation mask

data:image/png;base64,...
image_sizeobject

Original image dimensions

Example:
{ "width": 1920, "height": 1080 }

Complete Example

Request

{
  "model": "gambling-detection",
  "image_url": "https://example.com/image.jpg"
}

Response

{
  "success": true,
  "data": {
    "num_detections": 2,
    "detections": [
      {
        "bbox": {
          "x1": 100,
          "y1": 150,
          "x2": 300,
          "y2": 400
        },
        "score": 0.95,
        "class_id": 0,
        "class_name": "person",
        "mask_png": "data:image/png;base64,iVBORw0KGgo..."
      },
      {
        "bbox": {
          "x1": 400,
          "y1": 200,
          "x2": 600,
          "y2": 450
        },
        "score": 0.87,
        "class_id": 1,
        "class_name": "car",
        "mask_png": "data:image/png;base64,iVBORw0KGgo..."
      }
    ],
    "image_size": {
      "width": 1920,
      "height": 1080
    }
  }
}

Additional Information

Rate Limiting
If we throttle your request, you will receive a 429 HTTP error code along with an error message. You should then retry with an exponential back-off strategy, meaning that you should retry after 4 seconds, then 8 seconds, then 16 seconds, etc.
Supported Formats
gif, jpeg, jpg, png, webp
Maximum File Size
20MB
Tags:gamblingage-restrictedsegmentation

Ready to get started?

Integrate Gambling Detector into your application today with our easy-to-use API.

Learn more about image moderationBrowse all models