图像压缩技术
我对适用于图像的压缩技术的信息感兴趣，解压缩后不需要精确再现数据，只要结果看起来像图像一样即可。例如，我听说分形技术可以比标准数据压缩技术实现更好的图像压缩。有人可以指点我一些参考资料来查找吗？对于典型图像的不同分辨率，您能获得多好的压缩效果？谢谢。
I'm interested on information on compression techniques that are suitable for images, where exact reproduction of the data is not necessary after decompression, as long as the result looks ok as an image. I've heard that there are, for example, fractal techniques that achieve much better compression on images than standard data compression techniques. Can anyone point me to some references to look up? How good compression can you get for different resolutions in typical images? Thanks.
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（3）
好吧，今天发布的这篇文章比较了几种开源图像压缩实现：
https://extrememoderate.wordpress.com/2011/11/28/a-web-centric-image-compression-benchmark/
至于算法，我建议您开始花一些时间在维基百科上，那里有一些不错的文章。
至少，在开始寻找更多内容之前，您应该具备 DCT 和小波理论的一些基础知识：
http:// /en.wikipedia.org/wiki/Discrete_cosine_transform
http://en.wikipedia.org/wiki/Wavelet_transform#Wavelet_compression
Well, it just happen that this article has been published today comparing several open-source image compression implementations :
https://extrememoderate.wordpress.com/2011/11/28/a-web-centric-image-compression-benchmark/
As to algorithms, i would recommend you to start spending some time on Wikipedia, where a few good articles are available.
At the very least, you should have some basics in DCT and Wavelets theory before starting to look for more :
http://en.wikipedia.org/wiki/Discrete_cosine_transform
http://en.wikipedia.org/wiki/Wavelet_transform#Wavelet_compression
几年前，当我从事小波差分缩减工作时，我写了一些博客。这可能是一本有用的读物。我真的希望 OpenCL 能够支持编译 C++ 代码，因为我希望将其直接转换为 OpenCL，因为它是一种完美的可并行算法。
I wrote a bit of a blog when I was working on Wavelet Difference Reduction a few years back. It might be a useful read. I'm really wishing OpenCL would support compiling C++ code because I'd love to convert it directly to OpenCL as its a beautifully parallelisable algorithm.
一种易于实现的替代方案是块截断压缩，Anton Kruger 的文章“块截断压缩”对此进行了解释，该文章发表在“Dr. Dobb's Journal”的 XXX 期上：
http://drdobbs.com/database/184408745?pgno=26
（请注意，本文在 DDJ 的网页上的格式非常糟糕。可能值得简单地购买他们的 DVD，以获得更好的效果复制。）
基本思想是将亮度限制为图像小块内的 2 个不同值，并用单个位表示这些值。
One alternative which is easy to implement is block truncation compression, which was explained in Anton Kruger's article, "Block Truncation Compression", published in the XXX issue of "Dr. Dobb's Journal":
http://drdobbs.com/database/184408745?pgno=26
(Note that this article is extremely poorly formatted on DDJ's Web page. It might be worth simply buying their DVD for a better copy.)
The basic idea is to limit luminance to 2 distinct values within small blocks of the image, and to represent those values with single bits.