binocular_distance_msT_binocular_distance_msBinocularDistanceMsBinocularDistanceMsbinocular_distance_ms双目距离多线扫描（算子）
名称
binocular_distance_msT_binocular_distance_msBinocularDistanceMsBinocularDistanceMsbinocular_distance_ms — 使用多扫描线优化计算校正后的立体图像对的距离值。
签名
binocular_distance_ms(ImageRect1, ImageRect2 : Distance, Score : CamParamRect1, CamParamRect2, RelPoseRect, MinDisparity, MaxDisparity, SurfaceSmoothing, EdgeSmoothing, GenParamName, GenParamValue : )
Herror T_binocular_distance_ms(const Hobject ImageRect1, const Hobject ImageRect2, Hobject* Distance, Hobject* Score, const Htuple CamParamRect1, const Htuple CamParamRect2, const Htuple RelPoseRect, const Htuple MinDisparity, const Htuple MaxDisparity, const Htuple SurfaceSmoothing, const Htuple EdgeSmoothing, const Htuple GenParamName, const Htuple GenParamValue)
void BinocularDistanceMs(const HObject& ImageRect1, const HObject& ImageRect2, HObject* Distance, HObject* Score, const HTuple& CamParamRect1, const HTuple& CamParamRect2, const HTuple& RelPoseRect, const HTuple& MinDisparity, const HTuple& MaxDisparity, const HTuple& SurfaceSmoothing, const HTuple& EdgeSmoothing, const HTuple& GenParamName, const HTuple& GenParamValue)
HImage HImage::BinocularDistanceMs(const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const HTuple& GenParamName, const HTuple& GenParamValue) const
HImage HImage::BinocularDistanceMs(const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const HString& GenParamName, const HString& GenParamValue) const
HImage HImage::BinocularDistanceMs(const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const char* GenParamName, const char* GenParamValue) const
HImage HImage::BinocularDistanceMs(const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const wchar_t* GenParamName, const wchar_t* GenParamValue) const
(
Windows only)
HImage HCamPar::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const HTuple& GenParamName, const HTuple& GenParamValue) const
HImage HCamPar::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const HString& GenParamName, const HString& GenParamValue) const
HImage HCamPar::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const char* GenParamName, const char* GenParamValue) const
HImage HCamPar::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect2, const HPose& RelPoseRect, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const wchar_t* GenParamName, const wchar_t* GenParamValue) const
(
Windows only)
HImage HPose::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const HTuple& GenParamName, const HTuple& GenParamValue) const
HImage HPose::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const HString& GenParamName, const HString& GenParamValue) const
HImage HPose::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const char* GenParamName, const char* GenParamValue) const
HImage HPose::BinocularDistanceMs(const HImage& ImageRect1, const HImage& ImageRect2, HImage* Score, const HCamPar& CamParamRect1, const HCamPar& CamParamRect2, Hlong MinDisparity, Hlong MaxDisparity, Hlong SurfaceSmoothing, Hlong EdgeSmoothing, const wchar_t* GenParamName, const wchar_t* GenParamValue) const
(
Windows only)
static void HOperatorSet.BinocularDistanceMs(HObject imageRect1, HObject imageRect2, out HObject distance, out HObject score, HTuple camParamRect1, HTuple camParamRect2, HTuple relPoseRect, HTuple minDisparity, HTuple maxDisparity, HTuple surfaceSmoothing, HTuple edgeSmoothing, HTuple genParamName, HTuple genParamValue)
HImage HImage.BinocularDistanceMs(HImage imageRect2, out HImage score, HCamPar camParamRect1, HCamPar camParamRect2, HPose relPoseRect, int minDisparity, int maxDisparity, int surfaceSmoothing, int edgeSmoothing, HTuple genParamName, HTuple genParamValue)
HImage HImage.BinocularDistanceMs(HImage imageRect2, out HImage score, HCamPar camParamRect1, HCamPar camParamRect2, HPose relPoseRect, int minDisparity, int maxDisparity, int surfaceSmoothing, int edgeSmoothing, string genParamName, string genParamValue)
HImage HCamPar.BinocularDistanceMs(HImage imageRect1, HImage imageRect2, out HImage score, HCamPar camParamRect2, HPose relPoseRect, int minDisparity, int maxDisparity, int surfaceSmoothing, int edgeSmoothing, HTuple genParamName, HTuple genParamValue)
HImage HCamPar.BinocularDistanceMs(HImage imageRect1, HImage imageRect2, out HImage score, HCamPar camParamRect2, HPose relPoseRect, int minDisparity, int maxDisparity, int surfaceSmoothing, int edgeSmoothing, string genParamName, string genParamValue)
HImage HPose.BinocularDistanceMs(HImage imageRect1, HImage imageRect2, out HImage score, HCamPar camParamRect1, HCamPar camParamRect2, int minDisparity, int maxDisparity, int surfaceSmoothing, int edgeSmoothing, HTuple genParamName, HTuple genParamValue)
HImage HPose.BinocularDistanceMs(HImage imageRect1, HImage imageRect2, out HImage score, HCamPar camParamRect1, HCamPar camParamRect2, int minDisparity, int maxDisparity, int surfaceSmoothing, int edgeSmoothing, string genParamName, string genParamValue)
def binocular_distance_ms(image_rect_1: HObject, image_rect_2: HObject, cam_param_rect_1: Sequence[Union[int, float, str]], cam_param_rect_2: Sequence[Union[int, float, str]], rel_pose_rect: Sequence[Union[int, float]], min_disparity: int, max_disparity: int, surface_smoothing: int, edge_smoothing: int, gen_param_name: MaybeSequence[str], gen_param_value: MaybeSequence[Union[str, int]]) -> Tuple[HObject, HObject]
描述
binocular_distance_msbinocular_distance_msBinocularDistanceMsBinocularDistanceMsBinocularDistanceMsbinocular_distance_ms 通过多扫描线优化计算校正立体图像对的距离值。该算子首先参照 binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMsbinocular_disparity_ms 的方法计算校正图像 ImageRect1ImageRect1ImageRect1ImageRect1imageRect1image_rect_1 与 ImageRect2ImageRect2ImageRect2ImageRect2imageRect2image_rect_2 之间的视差。随后将所得视差值转换为对应三维世界点到校正立体相机系统的距离值，此过程类似于
disparity_to_distancedisparity_to_distanceDisparityToDistanceDisparityToDistanceDisparityToDistancedisparity_to_distance。最终距离值以单通道图像 DistanceDistanceDistanceDistancedistancedistance 形式返回，其中每个灰度值代表对应三维世界点到立体相机系统的距离。
binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMsbinocular_disparity_ms 需要一张参考图像
ImageRect1ImageRect1ImageRect1ImageRect1imageRect1image_rect_1 和一张搜索图像 ImageRect2ImageRect2ImageRect2ImageRect2imageRect2image_rect_2 ，两者都必须经过校正，即对应像素必须具有相同的行坐标。如果违反此假设，可通过 calibrate_camerascalibrate_camerasCalibrateCamerasCalibrateCamerasCalibrateCamerascalibrate_cameras、gen_binocular_rectification_mapgen_binocular_rectification_mapGenBinocularRectificationMapGenBinocularRectificationMapGenBinocularRectificationMapgen_binocular_rectification_map 和 map_imagemap_imageMapImageMapImageMapImagemap_image 算子对图像进行校正。
要将视差转换为距离，必须指定校正相机机 1 的内部参数
CamParamRect1CamParamRect1CamParamRect1CamParamRect1camParamRect1cam_param_rect_1 、校正相机 2 的内部参数
CamParamRect2CamParamRect2CamParamRect2CamParamRect2camParamRect2cam_param_rect_2，以及相机间的相对姿态
RelPoseRectRelPoseRectRelPoseRectRelPoseRectrelPoseRectrel_pose_rect。相对姿态定义了从校正后的相机系统 2 到校正后的相机系统 1 的点变换。这些参数可通过 calibrate_camerascalibrate_camerasCalibrateCamerasCalibrateCamerasCalibrateCamerascalibrate_cameras 和
gen_binocular_rectification_mapgen_binocular_rectification_mapGenBinocularRectificationMapGenBinocularRectificationMapGenBinocularRectificationMapgen_binocular_rectification_map 算子获取。
其余参数的详细说明可参见 binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMsbinocular_disparity_ms 的文档。
注意
若使用配备远心镜头的相机，DistanceDistanceDistanceDistancedistancedistance 并非定义为某一点到相机的距离，而是定义为该点到由两台相机的 y 轴及其基线构成的平面的距离（参见
gen_binocular_rectification_mapgen_binocular_rectification_mapGenBinocularRectificationMapGenBinocularRectificationMapGenBinocularRectificationMapgen_binocular_rectification_map）。
对于混合类型的立体设置（即原始相机中一台为透视相机而另一台为远心相机的立体设置；参见
gen_binocular_rectification_mapgen_binocular_rectification_mapGenBinocularRectificationMapGenBinocularRectificationMapGenBinocularRectificationMapgen_binocular_rectification_map），两台相机的校正平面相对于物体的位置会导致非常反直觉的距离值。因此，
binocular_distance_msbinocular_distance_msBinocularDistanceMsBinocularDistanceMsBinocularDistanceMsbinocular_distance_ms 不支持混合类型的立体设置。对于混合类型的立体设置，请使用
reconstruct_surface_stereoreconstruct_surface_stereoReconstructSurfaceStereoReconstructSurfaceStereoReconstructSurfaceStereoreconstruct_surface_stereo，其中参考坐标系可任意选择。或者，也可以使用
binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMsbinocular_disparity_ms 和 disparity_image_to_xyzdisparity_image_to_xyzDisparityImageToXyzDisparityImageToXyzDisparityImageToXyzdisparity_image_to_xyz。
此外，同时包含配备和未配备超中心镜头的相机的立体设置不受支持。
执行信息
- 支持 OpenCL 计算设备。
- 多线程类型：可重入（与非独占算子并行运行）。
- 多线程作用域：全局（可从任何线程调用）。
- 在元组级别上自动并行化。
- 在内部数据级别上自动并行化。
参数
ImageRect1ImageRect1ImageRect1ImageRect1imageRect1image_rect_1 (输入对象) singlechannelimage → objectHImageHObjectHImageHobject (byte*) *允许用于计算设备
相机 1 的校正图像。
ImageRect2ImageRect2ImageRect2ImageRect2imageRect2image_rect_2 (输入对象) singlechannelimage → objectHImageHObjectHImageHobject (byte*) *允许用于计算设备
相机 2 的校正图像。
DistanceDistanceDistanceDistancedistancedistance (输出对象) singlechannelimage → objectHImageHObjectHImageHobject * (real)
距离图像。
ScoreScoreScoreScorescorescore (输出对象) singlechannelimage → objectHImageHObjectHImageHobject * (real)
计算视差的分数。
CamParamRect1CamParamRect1CamParamRect1CamParamRect1camParamRect1cam_param_rect_1 (输入控制) campar → HCamPar, HTupleSequence[Union[int, float, str]]HTupleHtuple (real / integer / string) (double / int / long / string) (double / Hlong / HString) (double / Hlong / char*)
校正相机 1 的内部相机参数。
CamParamRect2CamParamRect2CamParamRect2CamParamRect2camParamRect2cam_param_rect_2 (输入控制) campar → HCamPar, HTupleSequence[Union[int, float, str]]HTupleHtuple (real / integer / string) (double / int / long / string) (double / Hlong / HString) (double / Hlong / char*)
校正相机 2 的内部相机参数。
RelPoseRectRelPoseRectRelPoseRectRelPoseRectrelPoseRectrel_pose_rect (输入控制) pose → HPose, HTupleSequence[Union[int, float]]HTupleHtuple (real / integer) (double / int / long) (double / Hlong) (double / Hlong)
从校正后的相机 2 到校正后的相机 1 的点变换。
元素数量：
7
MinDisparityMinDisparityMinDisparityMinDisparityminDisparitymin_disparity (输入控制) integer → HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
预期视差的最小值。
默认值：
-30
值范围：
-32768
≤
MinDisparity
MinDisparity
MinDisparity
MinDisparity
minDisparity
min_disparity
≤
32768
限制：
MinDisparity <= MaxDisparity
MaxDisparityMaxDisparityMaxDisparityMaxDisparitymaxDisparitymax_disparity (输入控制) integer → HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
预期视差的最大值。
默认值：
30
值范围：
-32768
≤
MaxDisparity
MaxDisparity
MaxDisparity
MaxDisparity
maxDisparity
max_disparity
≤
32768
限制：
MinDisparity <= MaxDisparity
SurfaceSmoothingSurfaceSmoothingSurfaceSmoothingSurfaceSmoothingsurfaceSmoothingsurface_smoothing (输入控制) integer → HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
表面平滑处理。
默认值：
50
建议值：
20, 50, 100
限制：
SurfaceSmoothing >= 0
EdgeSmoothingEdgeSmoothingEdgeSmoothingEdgeSmoothingedgeSmoothingedge_smoothing (输入控制) integer → HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
边缘平滑处理。
默认值：
50
建议值：
20, 50, 100
限制：
EdgeSmoothing >= 0
GenParamNameGenParamNameGenParamNameGenParamNamegenParamNamegen_param_name (输入控制) attribute.name(-array) → HTupleMaybeSequence[str]HTupleHtuple (string) (string) (HString) (char*)
多扫描线算法的参数名称。
默认值：
[]
值列表：
'consistency_check'"consistency_check""consistency_check""consistency_check""consistency_check""consistency_check", 'disparity_offset'"disparity_offset""disparity_offset""disparity_offset""disparity_offset""disparity_offset", 'num_levels'"num_levels""num_levels""num_levels""num_levels""num_levels", 'similarity_measure'"similarity_measure""similarity_measure""similarity_measure""similarity_measure""similarity_measure", 'sub_disparity'"sub_disparity""sub_disparity""sub_disparity""sub_disparity""sub_disparity"
GenParamValueGenParamValueGenParamValueGenParamValuegenParamValuegen_param_value (输入控制) attribute.value(-array) → HTupleMaybeSequence[Union[str, int]]HTupleHtuple (string / integer) (string / int / long) (HString / Hlong) (char* / Hlong)
多扫描线算法的参数值。
默认值：
[]
建议值：
'census_dense'"census_dense""census_dense""census_dense""census_dense""census_dense", 'census_sparse'"census_sparse""census_sparse""census_sparse""census_sparse""census_sparse", 'true'"true""true""true""true""true", 'false'"false""false""false""false""false"
结果
如果参数值正确 binocular_distance_msbinocular_distance_msBinocularDistanceMsBinocularDistanceMsBinocularDistanceMsbinocular_distance_ms 返回值 2 ( H_MSG_TRUE )。当输入为空（没有可用输入图像）时，可设置行为通过算子 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>)set_system("no_object_result",<Result>)。如有必要，则抛出异常。
可能的前趋
map_imagemap_imageMapImageMapImageMapImagemap_image
可能的后继
thresholdthresholdThresholdThresholdThresholdthreshold
替代
binocular_distancebinocular_distanceBinocularDistanceBinocularDistanceBinocularDistancebinocular_distance,
binocular_distance_mgbinocular_distance_mgBinocularDistanceMgBinocularDistanceMgBinocularDistanceMgbinocular_distance_mg,
binocular_disparitybinocular_disparityBinocularDisparityBinocularDisparityBinocularDisparitybinocular_disparity,
binocular_disparity_mgbinocular_disparity_mgBinocularDisparityMgBinocularDisparityMgBinocularDisparityMgbinocular_disparity_mg,
binocular_disparity_msbinocular_disparity_msBinocularDisparityMsBinocularDisparityMsBinocularDisparityMsbinocular_disparity_ms
另见
map_imagemap_imageMapImageMapImageMapImagemap_image,
gen_binocular_rectification_mapgen_binocular_rectification_mapGenBinocularRectificationMapGenBinocularRectificationMapGenBinocularRectificationMapgen_binocular_rectification_map,
binocular_calibrationbinocular_calibrationBinocularCalibrationBinocularCalibrationBinocularCalibrationbinocular_calibration,
disparity_to_distancedisparity_to_distanceDisparityToDistanceDisparityToDistanceDisparityToDistancedisparity_to_distance,
distance_to_disparitydistance_to_disparityDistanceToDisparityDistanceToDisparityDistanceToDisparitydistance_to_disparity,
disparity_image_to_xyzdisparity_image_to_xyzDisparityImageToXyzDisparityImageToXyzDisparityImageToXyzdisparity_image_to_xyz
模块
三维计量