binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMs (Operator)

Name

binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMs — Compute the disparities of a rectified stereo image pair using multi-scanline optimization.

Signature

binocular_disparity_ms(ImageRect1, ImageRect2 : Disparity, Score : MinDisparity, MaxDisparity, SurfaceSmoothing, EdgeSmoothing, GenParamName, GenParamValue : )

Description

binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMs calculates the disparity between two rectified stereo images ImageRect1ImageRect1ImageRect1ImageRect1imageRect1 and ImageRect2ImageRect2ImageRect2ImageRect2imageRect2 using multi-scanline optimization. The resulting disparity image is returned in DisparityDisparityDisparityDisparitydisparity. In contrast to binocular_distance_msbinocular_distance_msBinocularDistanceMsBinocularDistanceMsBinocularDistanceMs, the results are not transformed into distance values.

For this task, the three operators binocular_disparitybinocular_disparityBinocularDisparityBinocularDisparityBinocularDisparity, binocular_disparity_mgbinocular_disparity_mgBinocularDisparityMgBinocularDisparityMgBinocularDisparityMg, and binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMs can be used. binocular_disparitybinocular_disparityBinocularDisparityBinocularDisparityBinocularDisparity returns robust results in regions of sufficient texture but fails where is none. binocular_disparity_mgbinocular_disparity_mgBinocularDisparityMgBinocularDisparityMgBinocularDisparityMg interpolates low-texture regions but blurs discontinuities. binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMs preserves discontinuities and interpolates partially.

binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMs requires a reference image ImageRect1ImageRect1ImageRect1ImageRect1imageRect1 and a search image ImageRect2ImageRect2ImageRect2ImageRect2imageRect2 which both must be rectified, i.e., corresponding pixels must have the same row coordinate. If this assumption is violated, the images can be rectified by using the operators calibrate_camerascalibrate_camerasCalibrateCamerasCalibrateCamerasCalibrateCameras, gen_binocular_rectification_mapgen_binocular_rectification_mapGenBinocularRectificationMapGenBinocularRectificationMapGenBinocularRectificationMap, and map_imagemap_imageMapImageMapImageMapImage.

ImageRect1ImageRect1ImageRect1ImageRect1imageRect1 and ImageRect2ImageRect2ImageRect2ImageRect2imageRect2 can have different widths, but must have the same height. Given a pixel in ImageRect1ImageRect1ImageRect1ImageRect1imageRect1, the homologous pixel in ImageRect2ImageRect2ImageRect2ImageRect2imageRect2 is selected by searching along the corresponding row in ImageRect2ImageRect2ImageRect2ImageRect2imageRect2 and matching both pixels based on a similarity measure. The disparity is the number of pixels by which each pixel in ImageRect1ImageRect1ImageRect1ImageRect1imageRect1 needs to be moved to reach the homologous pixel in ImageRect2ImageRect2ImageRect2ImageRect2imageRect2. The search space is confined by the minimum and maximum disparity value MinDisparityMinDisparityMinDisparityMinDisparityminDisparity and MaxDisparityMaxDisparityMaxDisparityMaxDisparitymaxDisparity.

The similarity measure can be specified by setting GenParamNameGenParamNameGenParamNameGenParamNamegenParamName to 'similarity_measure'"similarity_measure""similarity_measure""similarity_measure""similarity_measure" and GenParamValueGenParamValueGenParamValueGenParamValuegenParamValue to the desired value. For both options 'census_dense'"census_dense""census_dense""census_dense""census_dense" (default) and 'census_sparse'"census_sparse""census_sparse""census_sparse""census_sparse", the similarity measure is based on the Census transform. A Census transformed image contains for every pixel information about the intensity topology within a support window around it. 'census_dense'"census_dense""census_dense""census_dense""census_dense" uses a dense 9 x 7 pixels window. 'census_sparse'"census_sparse""census_sparse""census_sparse""census_sparse" uses a sparse 15 x 15 pixels window where only a subset of the pixels is evaluated. While 'census_dense'"census_dense""census_dense""census_dense""census_dense" is more suitable for fine structures, 'census_sparse'"census_sparse""census_sparse""census_sparse""census_sparse" is more robust on low texture areas.

To calculate the disparities from the similarity measure, the intermediate results are optimized by a multi-scanline method. The optimization increases the robustness in low-texture areas without blurring discontinuities in the disparity image. The optimization is controlled by the parameters SurfaceSmoothingSurfaceSmoothingSurfaceSmoothingSurfaceSmoothingsurfaceSmoothing and EdgeSmoothingEdgeSmoothingEdgeSmoothingEdgeSmoothingedgeSmoothing. SurfaceSmoothingSurfaceSmoothingSurfaceSmoothingSurfaceSmoothingsurfaceSmoothing controls the smoothness within surfaces. High values suppress disparity differences of one pixel. EdgeSmoothingEdgeSmoothingEdgeSmoothingEdgeSmoothingedgeSmoothing controls the occurrence and the shape of edges. Low values allow many edges, high values lead to fewer and rounder edges. For both parameters, reasonable values usually range between 0 and 100. If both parameters are set to zero, no optimization is performed.

The resulting disparity is returned in the single-channel image DisparityDisparityDisparityDisparitydisparity. A quality measure for each disparity value is returned in ScoreScoreScoreScorescore, containing the best (lowest) result of the optimized similarity measure of a reference pixel.

Finally, there are two optional post-processing steps. A consistency check is applied if GenParamNameGenParamNameGenParamNameGenParamNamegenParamName is set to 'consistency_check'"consistency_check""consistency_check""consistency_check""consistency_check" and GenParamValueGenParamValueGenParamValueGenParamValuegenParamValue is set to 'true'"true""true""true""true" (default). This increases the robustness of the returned matches since the result relies on a concurrent direct and reverse match. It is switched off by setting GenParamValueGenParamValueGenParamValueGenParamValuegenParamValue to 'false'"false""false""false""false".

Sub-pixel refinement of disparities is performed by setting GenParamNameGenParamNameGenParamNameGenParamNamegenParamName to 'sub_disparity'"sub_disparity""sub_disparity""sub_disparity""sub_disparity" and GenParamValueGenParamValueGenParamValueGenParamValuegenParamValue to 'true'"true""true""true""true" (default). It is switched off by setting GenParamValueGenParamValueGenParamValueGenParamValuegenParamValue to 'false'"false""false""false""false".

The runtime of the operator is approximately linear to the image width, the image height, and the disparity range. Consequently, the disparity range should be chosen as narrow as possible for large images.

Execution Information

Supports OpenCL compute devices.
Multithreading type: reentrant (runs in parallel with non-exclusive operators).
Multithreading scope: global (may be called from any thread).
Automatically parallelized on tuple level.
Automatically parallelized on internal data level.

Parameters

ImageRect1ImageRect1ImageRect1ImageRect1imageRect1 (input_object) singlechannelimage → object (byte*) *allowed for compute devices

Rectified image of camera 1.

ImageRect2ImageRect2ImageRect2ImageRect2imageRect2 (input_object) singlechannelimage → object (byte*) *allowed for compute devices

Rectified image of camera 2.

DisparityDisparityDisparityDisparitydisparity (output_object) singlechannelimage → object (real)

Disparity map.

ScoreScoreScoreScorescore (output_object) singlechannelimage → object (real)

Score of the calculated disparity.

MinDisparityMinDisparityMinDisparityMinDisparityminDisparity (input_control) integer → (integer)

Minimum of the expected disparities.

Default value: -30

Typical range of values: -32768 ≤ MinDisparity MinDisparity MinDisparity MinDisparity minDisparity ≤ 32767

MaxDisparityMaxDisparityMaxDisparityMaxDisparitymaxDisparity (input_control) integer → (integer)

Maximum of the expected disparities.

Default value: 30

Typical range of values: -32768 ≤ MaxDisparity MaxDisparity MaxDisparity MaxDisparity maxDisparity ≤ 32767

Restriction: MinDisparity <= MaxDisparity

SurfaceSmoothingSurfaceSmoothingSurfaceSmoothingSurfaceSmoothingsurfaceSmoothing (input_control) integer → (integer)

Smoothing of surfaces.

Default value: 50

Suggested values: 20, 50, 100

Restriction: SurfaceSmoothing >= 0

EdgeSmoothingEdgeSmoothingEdgeSmoothingEdgeSmoothingedgeSmoothing (input_control) integer → (integer)

Smoothing of edges.

Default value: 50

Suggested values: 20, 50, 100

Restriction: EdgeSmoothing >= 0

GenParamNameGenParamNameGenParamNameGenParamNamegenParamName (input_control) attribute.name(-array) → (string)

Parameter name(s) for the multi-scanline algorithm.

Default value: []

List of values: 'consistency_check'"consistency_check""consistency_check""consistency_check""consistency_check", 'similarity_measure'"similarity_measure""similarity_measure""similarity_measure""similarity_measure", 'sub_disparity'"sub_disparity""sub_disparity""sub_disparity""sub_disparity"

GenParamValueGenParamValueGenParamValueGenParamValuegenParamValue (input_control) attribute.value(-array) → (string)

Parameter value(s) for the multi-scanline algorithm.

Default value: []

Suggested values: 'census_dense'"census_dense""census_dense""census_dense""census_dense", 'census_sparse'"census_sparse""census_sparse""census_sparse""census_sparse", 'true'"true""true""true""true", 'false'"false""false""false""false"

Example (HDevelop)

read_image (BaseballL, 'stereo/epipolar/baseball_l')
read_image (BaseballR, 'stereo/epipolar/baseball_r')
binocular_disparity_ms (BaseballL, BaseballR, Disparity, Score, \
                        -40, -10, 50, 50, [], [])

Result

If the parameter values are correct, binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMs returns the value 2 (H_MSG_TRUE). If the input is empty (no input images are available) the behavior can be set via set_system('no_object_result',<Result>)set_system("no_object_result",<Result>)SetSystem("no_object_result",<Result>)SetSystem("no_object_result",<Result>)SetSystem("no_object_result",<Result>). If necessary, an exception is raised.