Process Images¶

Note: The image redaction API is experimental and is not included in Redactor by default. Contact Sighthound to gain access.

HTTP Request¶

POST /api/v1/images:process

Request Body¶

Fields
`inputUri` ^Required	^string URI which points to a single image for input
`features` ^Required	Items Enum: `PERSON_DETECTION`, `HEAD_DETECTION`, `FACE_DETECTION`, `VEHICLE_DETECTION`, `LICENSE_PLATE_DETECTION`, `ID_DETECTION`, `SCREEN_DETECTION`, `DOCUMENT_DETECTION`, `MEDIA_RENDERING` String array of the desired processing features `PERSON_DETECTION` - Detect people `HEAD_DETECTION` - Detect heads (v6+) `FACE_DETECTION` - Detect faces `VEHICLE_DETECTION` - Detect vehicles `LICENSE_PLATE_DETECTION` - Detect vehicle license plates `ID_DETECTION` - Detect identification like driver's licenses, passports, and other IDs (v7+) `SCREEN_DETECTION` - Detect screens like monitors, laptops, and phones (v7+) `DOCUMENT_DETECTION` - Detect documents and other papers (v7+) `MEDIA_RENDERING` - Render redacted image and/or metadata. Required if output is desired.
`outputUri` ^Required	^string The URI to use for output. If the `outputUri` ends in a trailing slash, the redacted image will be saved as `image.jpg` inside of that directory, overwriting any existing file with the same name. If the `outputUri` does not end with trailing slash, it will be used as the the filename for the redacted, output image.
`notifications`	^{Array of objects (NotificationConfig)} Notification webhook descriptor. Users can enable webhooks to receive start, progress, and completion messages for their operation.
`imageContext`	^{object (ImageContext)} Additional configuration options. All optional.
`tag`	^object User-defined object that is returned in every notification.

Types¶

NotificationConfig¶

Fields
`method`	^string `"HTTP_POST"` is currently the only option
`uri`	^string The storage bucket URI to use for output
`events`	^{Array of string} Items Enum: `"START"` `"PROGRESS"` `"COMPLETION"`

ImageContext¶

Fields
`renderConfig`	object (RenderConfig)

RenderConfig¶

Fields
`exportMetadata`	^boolean Default: `false` Experimental feature in v4.5.0+ which outputs a JSON file with the raw, detection data.
`imageQuality`	^{number [ 0 .. 1 ]} Default: `0.65` Image compression quality which impacts the size of output. 0 is the lowest quality, 1 is highest, and 0.65 is the default.

Response Body¶

A success response returns an object with an Operation name in the format:
projects/{project-id}/locations/{location-id}/operations/{operation-id}

RESPONSE SCHEMA:	application/json
`name`	string<operationName>

Samples¶

Request¶

{
  "inputUri": "http://example.com/path/to/image.jpg",
  "features": [
    "FACE_DETECTION",
    "LICENSE_PLATE_DETECTION",
    "MEDIA_RENDERING"
  ],
  "outputUri": "file:///path/to/output/image-redacted.jpg",
  "notifications": [
    {
      "method": "HTTP_POST",
      "uri": "https://myhost/mywebhook/endpoint",
      "events": [
        "COMPLETION"
      ]
    }
  ],
  "imageContext": {
    "renderConfig": {
      "exportMetadata": true
    }
  },
  "tag": {
    "myid": 1607111026726
  }
}

Response¶

{
  "name": "projects/{project-id}/locations/{location-id}/operations/{operation-id}"
}