8

I have questions about both and

First of all, I'm using SIFT to identify similar image in real-time service. Like pictures by phone-camera, small amount of rotation and blurred effect could be.

And I found Phash. So, I test phash on its demo page. But result made me to sigh.

This is result of above test:

Demo of Phash

In this test, two images are fixed on x-axis. So they don' t have rotation. But right images' logo were removed and person was moved to left side. In my eye, This is 'Very Similar'. In addition, SIFT catch this completely.

Now, This is question.

  1. pHash is faster than SIFT?
  2. Is pHash's accuracy reliable?
  3. SIFT's output was too big to use in real-time service. So I must use hash to make output smaller size like LSH(Locality-sensitive hashing). Any other way to I try?
4

1 回答 1

9

好，我知道了。

pHash 无法将旋转和临界运动识别为同一事物。

在数据空间的情况下，pHash 非常适合使用。它的尺寸非常小：一个图像到一个哈希。然而，SIFT 需要 128 个字节来获得特征点。并且一张图像中有很多特征点。

最终，SIFT 可以比 pHash 更好地识别相似图像。但是需要越来越大的尺寸。

在速度台上，我还不能测试。但我认为，pHash 比 SIFT 更快，因为 SIFT 必须在一张图像上处理许多特征。

如果您对上述问题有其他答案，请告诉我。

于 2013-05-24T07:06:13.117 回答