Image Detection

High-precision object detection powered by state-of-the-art vision models. Provide an image URL and entity list to receive precise bounding box coordinates, annotated images, and accurate counts.

Examples

Input	Output	Input	Output
Cats	Detecting cats	Shopping cart	Detecting onions
Product listings	Detecting product listings	Police chase	Detecting police cars

What it does

This Actor analyzes images and finds objects you specify. It returns:

• Detection coordinates — Precise bounding box locations for each detected object
• Annotated image — Visual output with bounding boxes drawn on the original image

Input

Field	Type	Required	Description
imageUrl	string	Yes	URL of the image to analyze
categories	string[]	Yes	Objects to detect (e.g., ["cat", "dog", "person"])

Example

{
    "imageUrl": "https://media.hswstatic.com/eyJidWNrZXQiOiJjb250ZW50Lmhzd3N0YXRpYy5jb20iLCJrZXkiOiJnaWZcL3NodXR0ZXJzdG9jay0yMjc4Nzc2MTg3LWhlcm8uanBnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo4Mjh9fX0=",
    "categories": ["cat"]
}

Output

Dataset

Each run produces a result with:

Field	Description
imageUrl	The analyzed image URL
categories	Categories you searched for
detections	Array of detected objects with labels and bounding boxes
detectionSummary	Human-readable summary of what was found
annotatedImageKey	Key to download the annotated image
annotatedImageContentType	MIME type of the annotated image

Example Output

{
    "imageUrl": "https://media.hswstatic.com/eyJidWNrZXQiOiJjb250ZW50Lmhzd3N0YXRpYy5jb20iLCJrZXkiOiJnaWZcL3NodXR0ZXJzdG9jay0yMjc4Nzc2MTg3LWhlcm8uanBnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo4Mjh9fX0=",
    "categories": ["cat"],
    "detections": [
        {
            "label": "cat",
            "count": 3,
            "boxes": [
                {"x1": 46, "y1": 140, "x2": 401, "y2": 984},
                {"x1": 305, "y1": 3, "x2": 658, "y2": 988},
                {"x1": 640, "y1": 306, "x2": 934, "y2": 985}
            ]
        }
    ],
    "detectionSummary": "- cat: 3 instances detected",
    "annotatedImageKey": "annotated-image-1766024606223.png",
    "annotatedImageContentType": "image/png"
}

Key-Value Store

The annotated image with bounding boxes is saved to the key-value store with a unique timestamped key. Access it via the annotatedImageKey from the dataset output.

Use Cases

• E-commerce — Detect products in user-uploaded images
• Content moderation — Find specific objects in photos
• Inventory management — Count and locate items in warehouse images
• Wildlife monitoring — Detect animals in camera trap images
• Quality control — Identify defects or specific components in manufacturing images

Integrations

Connect this Actor with other tools using Apify integrations:

• Trigger detection from Zapier, Make, or n8n workflows
• Store results in Google Sheets, Airtable, or databases
• Send notifications when specific objects are detected