Halcon3D 点云染色（蒙皮）

根据halcon例程 Moments_Object_Model_3D

可以设置“lut”，“color_attrib”等参数来使点云根据某个方向上的尺度显示ColorSpace里的颜色

* Set visualization parameters for the 3d object model
* A lookup table (lut) is set that transforms attribute values into
* a color. In this example the color encodes the value of the
* z-component of the data. The alpha parameter defines the
* transparency of the object model.
GenParamName := ['lut','color_attrib','light_position','disp_pose','alpha']
GenParamValue := ['color1','coord_z','0.0 0.0 -0.3 1.0','true',0.9]

同样，可以使用

xyz_attrib_to_object_model_3d

这个算子在生成点云的时候，添加亮度图进3D模型中

xyz_attrib_to_object_model_3d (X, Y, Z, ImageRGB, ['&red', '&green', '&blue'], ObjectModel3D)
prepare_object_model_3d (ObjectModel3D, 'segmentation', 'true', [], [])
visualize_object_model_3d (WindowHandle, ObjectModel3D, [], [], ['red_channel_attrib', 'green_channel_attrib', 'blue_channel_attrib'], ['&red', '&green', '&blue'], [], [], [], PoseOut)

生成的Object 3D 包含了一张单通道的灰度图

可以通过Intensity参数来显示亮度图

GenParamName := ['light_position','disp_pose','alpha','intensity','point_size']
GenParamValue := ['0.0 0.0 -0.3 1.0','true',0.9,'&gray',1]
sample_object_model_3d (ObjectModel3D, 'fast', 0.05, [], [], SampledObjectModel3D)
visualize_object_model_3d (WindowHandle, SampledObjectModel3D, [], [], GenParamName, GenParamValue, [], [], [], PoseOut)

在.NET中略微有些区别，因为 xyz_attrib_to_object_model_3d 这是个halcon函数，并不是一算子；

仔细看了看这个函数，发现以前很多语句都是防呆的，比较重要的语句就是：

所以换到NET中的代码则为

                PCObjectModel3D.XyzToObjectModel3d(X,Y,DeepImage);
                //使用亮度图着色 
                HRegion DomainRegion = IntensityImage.GetDomain();
                HTuple Rows, Cols;
                DomainRegion.GetRegionPoints(out Rows,out Cols);
                double[] val =  IntensityImage.GetGrayval(Rows, Cols);
                PCObjectModel3D.SetObjectModel3dAttribMod("&amp;gray","points", val);
                InfoWindow.ShowMessage("亮度图着色 完成");

根据以上步骤：

生成点云并根据Z方向渲染颜色显示

而需要贴上亮度图显示的话，可以在显示参数中加上'intensity'-'&gray'这对参数（详细代码在上面已经贴出来了）

根据上面对比可以看出，根据Z方向着色可以很好的看出高度差异，但是产品表面的纹路看不清；

而贴图是亮度图的话可以很好的看出产品表面的反光情况，但是高度上不明显不直观；

根据Keyence提供的halcon代码，有一个折中的办法：

*将产品的高度信息转换成HSV颜色通道中的H通道色调

*手动创建一张单通道图作为饱和度通道 S通道

*将亮度图作为明度通道 V通道

*将HSV图片转成RGB图片并把RGB三个通道分别添加到模型中去

           double min_Color = 255;
            double max_Color = 0;
            double gray_mix_percent = 0.7;

            double DeepMin = 0;
            double DeepMax = 30;

            double Convert_K = (max_Color - min_Color) / (DeepMax - DeepMin);
            double Convert_B = min_Color - Convert_K * DeepMin;

            HImage HSV_H = DeepImage.ScaleImage(Convert_K, Convert_B);
            HSV_H = HSV_H.ConvertImageType("byte");
            //HSV_H.WriteImage("tiff",0,"D:\\TImage.tiff");

            //HRegion ZeroRegion = IntensityImage.Threshold(0.0, 0.0);
            //HRegion FullRegion = IntensityImage.GetDomain();
            //HRegion ResultRegion = FullRegion.Difference(ZeroRegion);

            //HImage IntensityBlobImage = IntensityImage.ReduceDomain(ResultRegion);
            HImage HSV_V = IntensityImage.ScaleImage(gray_mix_percent, 255 * (1 - gray_mix_percent));
            //HSV_V.WriteImage("tiff", 0, "D:\\TImageIntensity.tiff");

            int Imagewidth, imageheight;
            IntensityImage.GetImageSize(out Imagewidth,out imageheight);

            double Saturation = (1 - gray_mix_percent) * 512;
            HImage HSV_S = new HImage();
            HSV_S.GenImageSurfaceFirstOrder("byte",0,0, Saturation,0,0, Imagewidth, imageheight);

            HImage ChannelRed =  HSV_H.TransToRgb(HSV_S,HSV_V,out HImage ChannelGreen,out HImage ChannelBlue,"hsv");
            RainBowImage =  ChannelRed.Compose3(ChannelGreen, ChannelBlue);

                //使用彩虹图着色
                HImage Cred =  RainBowImage.AccessChannel(1);
                HRegion DomainRegion = Cred.GetDomain();
                HTuple Rows, Cols;
                DomainRegion.GetRegionPoints(out Rows, out Cols);
                double[] val = Cred.GetGrayval(Rows, Cols);
                PCObjectModel3D.SetObjectModel3dAttribMod("&amp;red", "points", val);

                HImage Cgreen = RainBowImage.AccessChannel(2);
                val = Cgreen.GetGrayval(Rows, Cols);
                PCObjectModel3D.SetObjectModel3dAttribMod("&amp;green", "points", val);

                HImage Cblue = RainBowImage.AccessChannel(3);
                val = Cblue.GetGrayval(Rows, Cols);
                PCObjectModel3D.SetObjectModel3dAttribMod("&amp;blue", "points", val);
                InfoWindow.ShowMessage("彩虹图着色 完成");

                //彩虹图着色显示
                HTuple paraname = new HTuple( "disp_pose", "alpha", "point_size", "disp_normals", "red_channel_attrib", "green_channel_attrib", "blue_channel_attrib");
                HTuple paraValue = new HTuple("true", 1, 1, "false","&amp;red", "&amp;green", "&amp;blue");

彩虹图效果，调整饱和度通道，可以显示更偏明度通道一些还是更偏色调通道一些。