图像移动校准
我有一系列在一段时间内拍摄的几乎相同的图像。然而，图像中的物体会随着时间的推移而漂移，我想纠正这一点。这样做有什么好处呢？
[编辑] 好吧，我可能必须解释一下为什么我要这么做。我拍摄了一系列不同 X 射线能量下物体的 X 射线图像。我现在想比较该物体的各种能量，但由于它会漂移，我必须首先纠正漂移。该物体没有尖锐的边缘或任何易于用于对齐的东西。因此我正在寻找一种更通用的方法
I have a series of mostly identical images taken over a period of time. However, the objects in the images drifts over time, and I would like to correct for this. What is a good was to do this?
[EDIT] Okay, I may have to explain why I'm going this. I've taken some series of X-ray images of objects at different X-ray energies. I now want to compare the object are the various energies, but since it drifts I have to correct for the drift first. The object has no sharps edges or anything which otherwise would be easy to use for alignment. Therefore I'm looking for a more general method
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
就其一般形式而言，这个问题被称为图像配准，并且是该领域的一个重要研究主题。图像处理社区。有多种不同的方法和算法，通常专门用于图像模态。根据您的图像，执行此操作可能很容易，也可能很困难。我建议使用文件交换中找到的注册方法之一。
根据您对图像的描述，似乎 刚性转换 应该足够了。在这种情况下，此方法应该工作顺利。
In its general form this problem is known as image registration, and is a large topic of research in the image processing community. There are a varity of different methods and algorithms, often specialized for image modality. Depending on your images, to do this could be easy, or it could be difficult. I would recommend using one of the registration methods found in the file-exchange.
Based on your description of your images, it seems a rigid transformation should be enough. In that case, this method should work nicely.