VTK 合并mask（指定label）

最新推荐文章于 2023-09-26 16:35:38 发布

ericohe

最新推荐文章于 2023-09-26 16:35:38 发布

阅读量545

点赞数

分类专栏： VTK C++ DICOM

本文链接：https://blog.csdn.net/Ericohe/article/details/116063515

版权

C++ 同时被 3 个专栏收录

69 篇文章 0 订阅

订阅专栏

VTK

40 篇文章 7 订阅

订阅专栏

DICOM

30 篇文章 1 订阅

订阅专栏

这段代码展示了如何使用VTK库将多个二维图像数据（mask）合并到一个三维图像中。首先，它获取第一个mask的数据维度并设置为合并后的mask的尺寸。然后，检查后续mask的尺寸是否匹配，如果不匹配则抛出错误。接着，遍历所有mask，将非零像素值替换为对应的label值，从而完成合并过程。

摘要由CSDN通过智能技术生成

void mergeMasks(Mask2Merged mask, vtkSmartPointer<vtkImageData>& mergedMask)
{
    mergedMask = vtkSmartPointer<vtkImageData>::New();
    int dimensions[3] = { 0 };
    int fullsize = 0;
    unsigned char c = mask.label;
    for (int i = 0; i < mask.maskPaths.size(); ++i)
    {
        vtkSmartPointer<vtkImageData> maskData = nullptr;
        loadImageData(mask.maskPaths[i].c_str(), maskData);
        if (0 == i)
        {
            mergedMask->SetOrigin(maskData->GetOrigin());
            mergedMask->SetSpacing(maskData->GetSpacing());
            mergedMask->SetDimensions(maskData->GetDimensions());
            mergedMask->AllocateScalars(VTK_UNSIGNED_CHAR, 1);
            maskData->GetDimensions(dimensions);
            auto mergedMaskPtr = (unsigned char*)mergedMask->GetScalarPointer();
            fullsize = dimensions[0] * dimensions[1] * dimensions[2];
            memset(mergedMaskPtr, 0, fullsize);
        }
        else
        {
            for (int m = 0; m < 3; m++)
            {
                if (dimensions[m] != maskData->GetDimensions()[m])
                {
                    std::cout << mask.maskPaths[i] << ", dimension is not correct!";
                    return;
                }
            }
        }

        auto maskPtr = (unsigned char*)maskData->GetScalarPointer();
        auto mergedMaskPtr = (unsigned char*)mergedMask->GetScalarPointer();
        for (int m = 0; m < fullsize; ++m)
        {
            if (maskPtr[m] > 0)
            {
                mergedMaskPtr[m] = c;
            }
        }
    }
}