Computer Vision API

Extract rich information from images to categorise and process visual data, with machine-assisted moderation of images to help curate your services.

Analyse an image

This feature returns information about visual content found in an image. Use tagging, descriptions and domain-specific models to identify content and label it with confidence. Apply the adult/racy settings to enable automated restriction of adult content. Identify image types and colour schemes in pictures.

See it in action

Gender Male
Age 36
Feature name: Value
Description { "Tags": [ "water", "swimming", "sport", "pool", "person", "man", "frisbee", "ocean", "blue", "bird", "riding", "top", "standing", "wave", "young", "body", "large", "game", "glass", "pond", "playing", "board", "catch", "clear", "boat", "white" ], "Captions": [ { "Text": "a man swimming in a pool of water", "Confidence": 0.8909298 } ] }
Tags [ { "Name": "water", "Confidence": 0.9997857 }, { "Name": "swimming", "Confidence": 0.955619633 }, { "Name": "sport", "Confidence": 0.953807831 }, { "Name": "pool", "Confidence": 0.9515978 }, { "Name": "person", "Confidence": 0.889862537 }, { "Name": "water sport", "Confidence": 0.664259 } ]
Image format "Jpeg"
Image dimensions 462 x 600
Clip art type 0
Line drawing type 0
Black and white false
Adult content false
Adult score 0.07518345
Racy false
Racy score 0.1814024
Categories [ { "Name": "people_swimming", "Score": 0.98046875 } ]
Faces [ { "Age": 36, "Gender": "Male", "FaceRectangle": { "Top": 133, "Left": 298, "Width": 121, "Height": 121 } } ]
Dominant colour background
"White"
Dominant colour foreground
"Grey"
Accent colour
#19A4B2

Want to build this?

Read text in images

Optical character recognition (OCR) detects text in an image and extract the recognised words into a machine-readable character stream. Analyse images to detect embedded text, generate character streams and enable searching. Take photos of text instead of copying to save time and effort.

See it in action

  1. Preview
  2. JSON

IF WE DID

ALL

THE THINGS

WE ARE

CAPABLÉ•

OF DOING,

WE WOULD

LITERALLY

ASTOUND

QURSELV*S.

{
  "TextAngle": 0.0,
  "Orientation": "NotDetected",
  "Language": "en",
  "Regions": [
    {
      "BoundingBox": "316,47,284,340",
      "Lines": [
        {
          "BoundingBox": "319,47,182,24",
          "Words": [
            {
              "BoundingBox": "319,47,42,24",
              "Text": "IF"
            },
            {
              "BoundingBox": "375,47,44,24",
              "Text": "WE"
            },
            {
              "BoundingBox": "435,47,66,23",
              "Text": "DID"
            }
          ]
        },
        {
          "BoundingBox": "316,74,204,69",
          "Words": [
            {
              "BoundingBox": "316,74,204,69",
              "Text": "ALL"
            }
          ]
        },
        {
          "BoundingBox": "318,147,207,24",
          "Words": [
            {
              "BoundingBox": "318,147,63,24",
              "Text": "THE"
            },
            {
              "BoundingBox": "397,147,128,24",
              "Text": "THINGS"
            }
          ]
        },
        {
          "BoundingBox": "316,176,125,23",
          "Words": [
            {
              "BoundingBox": "316,176,44,23",
              "Text": "WE"
            },
            {
              "BoundingBox": "375,176,66,23",
              "Text": "ARE"
            }
          ]
        },
        {
          "BoundingBox": "319,194,281,44",
          "Words": [
            {
              "BoundingBox": "319,194,281,44",
              "Text": "CAPABLÉ•"
            }
          ]
        },
        {
          "BoundingBox": "318,243,181,29",
          "Words": [
            {
              "BoundingBox": "318,243,43,23",
              "Text": "OF"
            },
            {
              "BoundingBox": "376,243,123,29",
              "Text": "DOING,"
            }
          ]
        },
        {
          "BoundingBox": "316,271,170,24",
          "Words": [
            {
              "BoundingBox": "316,272,44,23",
              "Text": "WE"
            },
            {
              "BoundingBox": "375,271,111,24",
              "Text": "WOULD"
            }
          ]
        },
        {
          "BoundingBox": "317,300,200,24",
          "Words": [
            {
              "BoundingBox": "317,300,200,24",
              "Text": "LITERALLY"
            }
          ]
        },
        {
          "BoundingBox": "316,328,157,24",
          "Words": [
            {
              "BoundingBox": "316,328,157,24",
              "Text": "ASTOUND"
            }
          ]
        },
        {
          "BoundingBox": "318,357,214,30",
          "Words": [
            {
              "BoundingBox": "318,357,214,30",
              "Text": "QURSELV*S."
            }
          ]
        }
      ]
    }
  ]
}

By uploading data for this demo, you agree that Microsoft may store it and use it to improve Microsoft services, such as making this API better. To help protect your privacy, we take steps to de-identify your data and keep it secure. We won’t publish your data or let other people use it.

Want to build this?

Preview: Read handwritten text from images

This technology (handwritten OCR) allows you to detect and extract handwritten text from notes, letters, essays, whiteboards, forms etc. It works with different surfaces and backgrounds, such as white paper, yellow sticky notes and whiteboards.

Handwritten text recognition saves time and effort and can make you more productive by allowing you to take images of text, rather than having to transcribe it. It makes it possible to digitise notes, which then allows you to implement quick and easy search. It also reduces paper clutter.

Note: this technology is currently in preview and is only available for English text.

To try this optical character recognition demo, upload a locally stored image or provide an image URL. We don’t store the images you supply for this demo unless you give us permission.

See it in action