首页>
外国专利>
Feature-based image processing using feature images extracted from different iterations
Feature-based image processing using feature images extracted from different iterations
展开▼
机译:使用从不同的迭代中提取的特征图像的基于特征的图像处理
展开▼
页面导航
摘要
著录项
相似文献
摘要
Image processing performed by a computer (22) includes iterative image reconstruction or refinement (26, 56) that produces a series of update images ending in an iteratively reconstructed or refined image. A difference image (34, 64) is computed between a first update image (30, 60) and a second update image (32, 62) of the series. The difference image is converted to a feature image (40) and is used in the iterative processing (26, 56) or in post-processing (44) performed on the iteratively reconstructed or refined images or images from different reconstruction or refinement techniques. In another embodiment, first and second image reconstructions (81, 83) are performed to generate respective first and second reconstructed images (80, 82). A difference image (84) is computed between two images each selected from the group: the first reconstructed image, an update image of the first reconstruction, the second reconstructed image, and an update image of the second reconstruction. A feature image is generated from the difference image and used to combine the first and second reconstructed images.
展开▼