Operator Reference
set_structured_light_model_param (Operator)
set_structured_light_model_param — Set parameters of a structured light model.
Signature
set_structured_light_model_param( : : StructuredLightModel, GenParamName, GenParamValue : )
Description
The operator set_structured_light_model_param is used to
manipulate the parameters of a structured light model
StructuredLightModel. The current parameter settings
can be queried with the operator get_structured_light_model_param.
For an explanation of the concept of structured light and its supported applications, see the introduction of chapter Inspection / Structured Light.
General parameters:
- 'persistence':
-
Enables (
GenParamValue='true') or disables (GenParamValue='false') the 'persistence' mode of the structured light model. When in persistence mode, the model stores intermediate results of the decoding, which can be inspected later byget_structured_light_object.Note that the model might need significant memory space in this mode. Therefore, we recommend to enable this mode only during the setup and for debugging a structured light model setup.
If 'persistence' is changed, all results of the model are cleared.
Values: 'true', 'false', 1, 0
Default: 'false'
Parameters which influence the appearance of the generated pattern images:
- 'pattern_width':
-
Sets the width of the pattern images that are generated while calling
gen_structured_light_pattern. Usually, the value should be set to the width of the monitor or projector that is used to display the pattern images.The value of 'pattern_width' has to be larger than the value 'min_stripe_width'/2. In case that 'pattern_type' is 'single_stripe', it also has to be larger than 'single_stripe_width'/2. In case that a 'camera_setup_model' is set in the structured light model, the value of 'pattern_width' has to match the width of the projector parameters stored in the camera setup model (at Index 1).
If 'pattern_width' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: integer values larger or equal to 1
Default: 1024
- 'pattern_height':
-
Sets the height of the pattern images that are generated while calling
gen_structured_light_pattern. Usually, the value should be set to the height of the monitor or projector that is used to display the pattern images.The value of 'pattern_height' has to be larger than the value 'min_stripe_width'/2. In case that 'pattern_type' is 'single_stripe', it also has to be larger than 'single_stripe_width'/2. In case that a 'camera_setup_model' is set in the structured light model, the value of 'pattern_height' has to match the height of the projector parameters stored in the camera setup model (at Index 1).
If 'pattern_height' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: integer values larger or equal to 1
Default: 1024
- 'min_stripe_width':
-
Sets the width (in pixels) of the finest stripe of the pattern images that are generated while calling
gen_structured_light_pattern.The value has to be a power of two. Furthermore, it has to be smaller or equal the minimum of the parameters 'pattern_width' and 'pattern_height' after rounding both up to their next power of two. In case that 'pattern_type' is 'single_stripe', it also has to be smaller or equal the 'single_stripe_width'.
If 'min_stripe_width' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: integer values larger or equal to 4 and a power of 2
Default: 32
- 'normalization':
-
Sets the normalization mode of the structured light model. If the parameter is set to 'global',
gen_structured_light_patterngenerates two additional pattern images composed of a black and a white image. These images are then used in the decoding process to determine the actual pattern region. If instead, the value 'inverted_pattern' is chosen, each Gray code image is additionally inverted and the pattern region is segmented iteratively using the Gray code images and their inverted version.If 'normalization' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: 'global', 'inverted_pattern'
Default: 'global'
- 'pattern_orientation':
-
Sets the orientation of the pattern images that are generated while calling
gen_structured_light_pattern. If the parameter is set to 'both', pattern images with vertical stripes as well as pattern images with horizontal stripes are created. For 'vertical' the pattern images consist of vertical stripes only, whereas for 'horizontal' only horizontal stripes are generated.If 'pattern_orientation' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: 'both', 'vertical', 'horizontal'
Default: 'both'
- 'pattern_type':
-
Sets the type of the pattern images that are generated while calling
gen_structured_light_pattern.'pattern_type' Generated Images 'gray_code' Gray code images 'gray_code_and_phase_shift' Gray code images and phase shift images 'single_stripe' Single stripe images, Gray code images and phase shift images Usually, 'gray_code_and_phase_shift' leads to more accurate results than 'gray_code' because the combination between Gray code and phase images yields sub-pixel precise monitor or projector coordinates. 'single_stripe' can increase robustness in case of only partially specular surfaces.
If 'pattern_type' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: 'gray_code_and_phase_shift', 'gray_code', 'single_stripe'
Default: 'gray_code_and_phase_shift'
- 'single_stripe_width':
-
Sets the width (in pixels) of the single stripe that is generated while calling
gen_structured_light_patternusing the 'pattern_type' 'single_stripe'.The value has to be a power of two and has to be greater or equal 'min_stripe_width'. Furthermore, it has to be smaller or equal the minimum of the parameters 'pattern_width' and 'pattern_height' after rounding both up to their next power of two.
Note that 'single_stripe_width' can only be set in case that 'pattern_type' is 'single_stripe'. When 'pattern_type' is changed to 'single_stripe', 'single_stripe_width' is set to some power of two roughly halfway between the currently set 'min_stripe_width' and the minimum of 'pattern_width' and 'pattern_height'.
If 'single_stripe_width' is changed, the model is not decodable anymore and all possibly available results are cleared. Therefore,
gen_structured_light_patternhas to be called before decoding the model usingdecode_structured_light_pattern.Values: integer values larger or equal to 4 and a power of 2
Default: power of two roughly halfway between the currently set 'min_stripe_width' and 'pattern_width', 'pattern_height'
Parameters which influence the segmentation of the pattern region:
- 'min_gray_difference':
-
Sets the minimum gray value difference for the Gray code images. This value is used within
decode_structured_light_patternto determine the binarized images. Thereby, for each pixel of the camera images the gray value difference between the lighted and the unlighted case is calculated. If the gray value difference of a pixel is smaller than 'min_gray_difference', it is excluded from the domain of the binarized images (seeget_structured_light_object) and thus also from further calculations.If 'min_gray_difference' is set to the value 0, the segmented region is identical to the input region of the camera images with the only exception that pixels with a decoding result outside the pattern images are excluded.
Values: integer values larger or equal to 0
Default: 30
Parameters for structured light models of type 'deflectometry':
- 'derivative_sigma':
-
Sets the sigma of the Gaussian (i.e. the amount of smoothing) that is used for the convolution of the correspondence image(s) to calculate the defect image.
Values: float or integer value larger or equal 0.01
Default: 2
Parameters for structured light models of type '3d_reconstruction':
- 'camera_setup_model':
-
Sets a copy of the calibrated camera setup model containing the calibration parameters of the structured light sensor, such as its camera and projector parameters and poses. Here, the projector of the sensor is modeled as an 'inverse camera'.
reconstruct_surface_structured_lightuses the calibration parameters to determine the reconstructed surface in world coordinates. Please refer to the example program on structured light calibration in order to obtain the calibrated camera setup model (for further details on calibration of multiple cameras see Calibration / Multi-View).The camera of the sensor has to correspond to Index 0 in the camera setup model, the projector of the sensor to Index 1. In order to delete a set camera setup model from
StructuredLightModel, HNULL has to be passed.Note that the width and height of the projector parameters stored in the camera setup model (at Index 1) have to match the set 'pattern_width' and 'pattern_height' of the structured light model. In subsequent calls to
decode_structured_light_patternandreconstruct_surface_structured_light, the width and height of the camera parameters stored in the camera setup model (at Index 0) have to match the dimensions of the input images ofdecode_structured_light_pattern.So far, only projectors are supported which can be modeled as an inverse pinhole camera, i.e., only projectors of type 'area_scan_division' and 'area_scan_polynomial'. For the camera, all area scan types except hypercentric ones are allowed.
If 'camera_setup_model' is changed, all possibly available results are cleared, so
decode_structured_light_patternhas to be called before reconstructing the model usingreconstruct_surface_structured_light.Default: HNULL
Execution Information
- Multithreading type: reentrant (runs in parallel with non-exclusive operators).
- Multithreading scope: global (may be called from any thread).
- Processed without parallelization.
This operator modifies the state of the following input parameter:
During execution of this operator, access to the value of this parameter must be synchronized if it is used across multiple threads.
Parameters
StructuredLightModel (input_control, state is modified) structured_light_model → (handle)
Handle of the structured light model.
GenParamName (input_control) attribute.name(-array) → (string)
Name of the model parameter to be adjusted.
Default: 'min_stripe_width'
List of values: 'camera_setup_model', 'derivative_sigma', 'min_gray_difference', 'min_stripe_width', 'normalization', 'pattern_height', 'pattern_orientation', 'pattern_type', 'pattern_width', 'persistence', 'single_stripe_width'
GenParamValue (input_control) attribute.value(-array) → (integer / real / string)
New value of the model parameter.
Default: 32
Suggested values: 0, 0.01, 0.5, 0.7, 1, 1.4, 5, 50.0, 64, 128, 256, 1024, 'true', 'false', 'both', 'vertical', 'horizontal', 'global', 'inverted_pattern', 'gray_code_and_phase_shift', 'gray_code', 'single_stripe'
Example (HDevelop)
* Create the model
create_structured_light_model ('deflectometry', StructuredLightModel)
* Set the size of the monitor
set_structured_light_model_param (StructuredLightModel, \
'pattern_width', 1600)
set_structured_light_model_param (StructuredLightModel, \
'pattern_height', 1200)
* Set the smallest width of the stripes in the pattern
set_structured_light_model_param (StructuredLightModel, \
'min_stripe_width', 8)
* Generate the patterns to project
gen_structured_light_pattern (PatternImages, StructuredLightModel)
* Set the expected black/white contrast in the region of interest
set_structured_light_model_param (StructuredLightModel, \
'min_gray_difference', 70)
* Decode the camera images
decode_structured_light_pattern (CameraImages, StructuredLightModel)
* Get the computed correspondences and defects
get_structured_light_object (CorrespondenceImages, StructuredLightModel, \
'correspondence_image')
set_structured_light_model_param (StructuredLightModel, 'derivative_sigma', \
Sigma)
get_structured_light_object (DefectImage, StructuredLightModel, \
'defect_image')
Result
The operator set_structured_light_model_param returns the value
2 (
H_MSG_TRUE)
if the given parameters are valid and within acceptable range.
Otherwise, an exception will be raised.
Possible Predecessors
See also
get_structured_light_model_param
Module
3D Metrology