Amazon Rekognition · Arazzo Workflow

Amazon Rekognition Text and Moderation Screen

Version 1.0.0

Extract text from an image and then screen the same image for unsafe content.

1 workflow 1 source API 1 provider
View Spec View on GitHub Celebrity RecognitionComputer VisionContent ModerationCustom LabelsDeep LearningFace LivenessFacial RecognitionImage AnalysisMachine LearningObject DetectionText DetectionVideo AnalysisArazzoWorkflows

Provider

amazon-rekognition

Workflows

text-and-moderation-screen
Detect text then detect moderation labels for the same image.
Extracts overlaid or embedded text from an image and screens the same image for unsafe content in a single pass.
2 steps inputs: bucket, minModerationConfidence, name outputs: moderationLabels, textDetections
1
detectText
detectText
Extract machine-readable text from the image.
2
detectModeration
detectModerationLabels
Screen the same image for unsafe or inappropriate content.

Source API Descriptions

Arazzo Workflow Specification

amazon-rekognition-text-and-moderation-screen-workflow.yml Raw ↑
arazzo: 1.0.1
info:
  title: Amazon Rekognition Text and Moderation Screen
  summary: Extract text from an image and then screen the same image for unsafe content.
  description: >-
    A combined OCR-plus-safety flow for Amazon Rekognition. The workflow runs
    DetectText to extract machine-readable text from an image and then runs
    DetectModerationLabels against the same image to flag unsafe content, which
    is useful for moderating image-based posts that also carry overlaid text.
    Each step spells out its AWS JSON 1.1 request inline, including the
    protocol-required X-Amz-Target header.
  version: 1.0.0
sourceDescriptions:
- name: rekognitionApi
  url: ../openapi/amazon-rekognition-openapi.yml
  type: openapi
workflows:
- workflowId: text-and-moderation-screen
  summary: Detect text then detect moderation labels for the same image.
  description: >-
    Extracts overlaid or embedded text from an image and screens the same image
    for unsafe content in a single pass.
  inputs:
    type: object
    required:
    - bucket
    - name
    properties:
      bucket:
        type: string
        description: S3 bucket holding the image to analyze.
      name:
        type: string
        description: S3 object key of the image.
      minModerationConfidence:
        type: number
        description: Minimum confidence for returned moderation labels.
        default: 50.0
  steps:
  - stepId: detectText
    description: Extract machine-readable text from the image.
    operationId: detectText
    parameters:
    - name: X-Amz-Target
      in: header
      value: RekognitionService.DetectText
    requestBody:
      contentType: application/x-amz-json-1.1
      payload:
        Image:
          S3Object:
            Bucket: $inputs.bucket
            Name: $inputs.name
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      textDetections: $response.body#/TextDetections
      textModelVersion: $response.body#/TextModelVersion
  - stepId: detectModeration
    description: Screen the same image for unsafe or inappropriate content.
    operationId: detectModerationLabels
    parameters:
    - name: X-Amz-Target
      in: header
      value: RekognitionService.DetectModerationLabels
    requestBody:
      contentType: application/x-amz-json-1.1
      payload:
        Image:
          S3Object:
            Bucket: $inputs.bucket
            Name: $inputs.name
        MinConfidence: $inputs.minModerationConfidence
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      moderationLabels: $response.body#/ModerationLabels
      moderationModelVersion: $response.body#/ModerationModelVersion
  outputs:
    textDetections: $steps.detectText.outputs.textDetections
    moderationLabels: $steps.detectModeration.outputs.moderationLabels