If you want to do feature extraction you should look more into computer vision libraries. Image processing is more in the of the kind of algorithms Photoshop does, i.e. to enhance/improve/transform images, not the retrieval of information out of them. The terms are sometimes uses synonymously though.
You might want to check out OpenCV, they provide some feature extractors and it's an easy to use library for C++.
For medical images, I strongly recommend ITK, which has been developed specifically for this purpose and is supported by major US government research funding agencies, such as the National Science Foundation, or the National Cancer Institute.
I used LTIlib for crop image processing. I know medicine is quite a little different, but the complexity of "understanding" aerial images is not trivial, trust me.
I will advocate LTIlib anyway for its "C++" flavour that fits very well. Performance is also pretty good. My (now) old 8 MPix images earned results with LTI in a comparablespeed with MVTec Halcon did.
In image processing you learn to enhance,increase contrast ,sharpness ,find specific shape in images, much like all the photo editor do. Feature extraction,stereo vision,object tracking is a part of Computer vision library .If you are working on medical images you will need both images processing toolbox and Computer vision toolbox .For this you can go with openCV .But a better option is to go with MATLAB toolbox which is easy to use.You will probabily need machine learning toolbox later in your research work ,for this matlab will provide a more complete solution.
发布评论
评论（4）
如果你想进行特征提取，你应该更多地研究计算机视觉库。图像处理更多地采用 Photoshop 的算法，即增强/改进/转换图像，而不是从中检索信息。但这些术语有时作为同义词使用。
您可能想查看 OpenCV，它们提供了一些特征提取器，并且它是一个易于使用的 C++ 库。
If you want to do feature extraction you should look more into computer vision libraries. Image processing is more in the of the kind of algorithms Photoshop does, i.e. to enhance/improve/transform images, not the retrieval of information out of them. The terms are sometimes uses synonymously though.
You might want to check out OpenCV, they provide some feature extractors and it's an easy to use library for C++.
对于医学图像，我强烈推荐ITK，它是专门为此目的开发的，并得到了美国主要政府的支持研究资助机构，例如国家科学基金会或国家癌症研究所。
For medical images, I strongly recommend ITK, which has been developed specifically for this purpose and is supported by major US government research funding agencies, such as the National Science Foundation, or the National Cancer Institute.
我使用 LTIlib 进行裁剪图像处理。
我知道医学有很大不同，但“理解”航拍图像的复杂性并非微不足道，相信我。
无论如何，我会提倡 LTIlib，因为它的“C++”风格非常适合。性能也相当不错。我（现在）旧的 8 MPix 图像通过 LTI 获得的结果与 MVTec Halcon 的速度相当。
I used LTIlib for crop image processing.
I know medicine is quite a little different, but the complexity of "understanding" aerial images is not trivial, trust me.
I will advocate LTIlib anyway for its "C++" flavour that fits very well. Performance is also pretty good. My (now) old 8 MPix images earned results with LTI in a comparablespeed with MVTec Halcon did.
在图像处理中，您学习增强、增加对比度、清晰度、查找图像中的特定形状，就像所有照片编辑器所做的那样。
特征提取、立体视觉、目标跟踪是计算机视觉库的一部分。如果您正在处理医学图像，您将需要图像处理工具箱和计算机视觉工具箱。为此，您可以使用 openCV 。但更好的选择是去带有易于使用的MATLAB工具箱。您在以后的研究工作中可能会需要机器学习工具箱，因为这个matlab将提供更完整的解决方案。
In image processing you learn to enhance,increase contrast ,sharpness ,find specific shape in images, much like all the photo editor do.
Feature extraction,stereo vision,object tracking is a part of Computer vision library .If you are working on medical images you will need both images processing toolbox and Computer vision toolbox .For this you can go with openCV .But a better option is to go with MATLAB toolbox which is easy to use.You will probabily need machine learning toolbox later in your research work ,for this matlab will provide a more complete solution.