Image Anonymization
Image Anonymization (Redact)
Images are processed with Tesseract OCR to extract text, then PII entities are detected and redacted with black rectangles covering the sensitive regions.
Bildanonymisierung
Anonimización de Imágenes
Anonymisation d'Images
Step-by-Step Process Description
Image anonymization uses Tesseract OCR (38 languages supported) to extract text regions from the image, then runs Presidio NER to classify PII entities and compute their bounding box coordinates. Detected regions are highlighted with colored overlays before redaction — you choose which entity types to redact before applying.
Image Anonymization — Step-by-step guide
- Navigate to the Image tab. The dropzone accepts PNG, JPG, BMP, and TIFF images up to 10 MB.
- Drop an image. The preview shows the image with visible PII. Configure the fill color (black, red, green, blue, gray), detection preset, and OCR language. Click "Analyze".
- The app runs Tesseract OCR to extract text, then Presidio NER to detect PII entities and their bounding box positions within the image.
- Review detected entities with colored bounding boxes overlaid on the image. Entity type filter badges (Person, Email, Phone, Location, Organization) with checkboxes allow selecting which to redact. Click "Redact Selected".
- Selected PII regions are covered with colored fill rectangles.
- Side-by-side before/after comparison: the original image alongside the redacted version with bars covering all detected PII. Use the Save button to download the redacted image.