|
Cpp-Processing
|
Configuration that describes processing of a particular class This structure contains the type of blurring to be applied to detections of a specific class. More...
#include <CommonParameters.h>
Public Attributes | |
| BlurType | blur_type {BlurType::BBox_Rectangle} |
| celantur::BlurType of blur to be applied to the detection. More... | |
| float | gradient_start {0.3} |
| Responsible for the starting point of the gradient blur. Larger values mean that the blur starts furhter away from the edge of the detection. For example, 0.0 will mean that blur will start exactly on the detection bounding box, while 0.5 will mean that blur will cover 1.5 times larger area than the detection bounding box. Value shoud be between 0.0 and 1.0. More... | |
| float | gradient_stop {0.0} |
| Responsible for the point where the blur is absolute. Value of 0.0 guarantees that the blur will be absolute at the edge of the detection bounding box and is the safest choice. However, if one finds consistently that the detection is too large compared to anonymised object, one can increase this value to make the blur reach absolute faster. Value should be between 0.0 and 1.0. More... | |
| float | kernel_size {0.2f} |
| Responsible for strength of the blur applied. Values between 0.0 and 1.0 are considered relative to the size of the detection. Values greater than 1.0 are considered absolute pixel values. Default is 0.2 (20% of the detection size). Values too small (below 0.1) might not produce sufficient anonymization, while values too large (above 0.5) might overly blur the image and lose important context. More... | |
| bool | beautify_blur {true} |
| If true, applies additional beautification to the blur effect, making it visually more appealing. If false, applies a straightforward blur without additional processing. Default is true. More... | |
Configuration that describes processing of a particular class This structure contains the type of blurring to be applied to detections of a specific class.
for usage see ProcessorParams and PerTypeProcessingConfig
| bool celantur::DetectionProcessingConfig::beautify_blur {true} |
If true, applies additional beautification to the blur effect, making it visually more appealing. If false, applies a straightforward blur without additional processing. Default is true.
| BlurType celantur::DetectionProcessingConfig::blur_type {BlurType::BBox_Rectangle} |
celantur::BlurType of blur to be applied to the detection.
| float celantur::DetectionProcessingConfig::gradient_start {0.3} |
Responsible for the starting point of the gradient blur. Larger values mean that the blur starts furhter away from the edge of the detection. For example, 0.0 will mean that blur will start exactly on the detection bounding box, while 0.5 will mean that blur will cover 1.5 times larger area than the detection bounding box. Value shoud be between 0.0 and 1.0.
| float celantur::DetectionProcessingConfig::gradient_stop {0.0} |
Responsible for the point where the blur is absolute. Value of 0.0 guarantees that the blur will be absolute at the edge of the detection bounding box and is the safest choice. However, if one finds consistently that the detection is too large compared to anonymised object, one can increase this value to make the blur reach absolute faster. Value should be between 0.0 and 1.0.
| float celantur::DetectionProcessingConfig::kernel_size {0.2f} |
Responsible for strength of the blur applied. Values between 0.0 and 1.0 are considered relative to the size of the detection. Values greater than 1.0 are considered absolute pixel values. Default is 0.2 (20% of the detection size). Values too small (below 0.1) might not produce sufficient anonymization, while values too large (above 0.5) might overly blur the image and lose important context.
1.9.1