使用opencv在iphone中进行图像匹配/检测
我有 2 个图像，即大图像和小图像。我想检测小图像是否存在于大图像中的任何位置，无论其方向或变换（旋转）如何。如果存在，则应返回 true，否则返回 false。我一直在尝试 openCV 中的模板匹配（cvMatchTemplate）方法，但还没有达到任何地方，因为真匹配和假匹配的输出之间没有太大区别。我的要求是否仍然可以使用 cvMatchTemplate 或者 openCV 中是否有其他方法可以实现此目的。
I have 2 images, say bigImage and smallImage. I want to detect whether the smallImage is there anywhere in the bigImage, irrespective of its orientation or transforms(rotations). If its there, it should return true and otherwise return false. I have been going through the template matching (cvMatchTemplate) method in openCV, but haven't reach anywhere since there aren't much difference between the output for a true and a false match. Is my requirement still possible using cvMatchTemplate or are there any other methods in openCV for getting to this.
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）
这是一个非常普遍且非常困难的问题，有很多解决方案。它们都不是灵丹妙药。希望其中之一能够解决您的特定问题。尝试看看 SIFT、颜色直方图匹配、SURF、HoG、Viola-Jones、DOT、ORB。其中一些具有 OpenCV 实现；其他人则不然。您应该阅读可用的选项及其优点和缺点，然后尝试一些。
This is a very general and very hard problem, with lots of solutions out there. None of them are silver bullets; hopefully, one of them will work for your particular problem. Try taking a look at SIFT, color histogram matching, SURF, HoG, Viola-Jones, DOT, ORB. Some of these have OpenCV implementations; others don't. You should read up on the available options and their strengths and weaknesses, and then try some out.