Expand description
Types used in image processing operations
This module defines various enums that represent different options and configurations for image processing operations in the OCR pipeline.
Structs§
- Image
Scale Info - Information about image scaling during preprocessing
Enums§
- BoxType
- Specifies the type of bounding box used for text detection
- Color
Order - Specifies whether the model expects RGB or BGR channel ordering.
- Crop
Mode - Specifies how to crop an image when the aspect ratios don’t match
- Limit
Type - Specifies how to limit the size of an image during resizing operations
- Resize
Type - Specifies different strategies for resizing images
- Score
Mode - Specifies the mode for calculating scores in text detection/recognition
- Tensor
Layout - Specifies the data layout of an image tensor