OpenCV学习笔记（七）：opencv保存png图片的压缩编码问题

今天在使用cv::imwrite保存PNG图像的时候发现了一个有趣的现象：

保存128*128分辨率16位的图像，PNG图像压缩参数CV_IMWRITE_PNG_COMPRESSION设置为0，得到大小为32.2kB（33012字节），计算每一个像素的值为128*128*16/8/1024=32kB，算上文件头等信息是刚刚好的；

保存256*256分辨率16位的图像，PNG图像压缩参数CV_IMWRITE_PNG_COMPRESSION设置为0，得到大小为22.3kB（22934字节），计算每一个像素的值为256*256*16/8/1024=128kB，算上文件头等信息应该是128kB多一点。

选择的参数是一样的，图像容量应该变为4倍才是正常的，为什么反而变小了呢？还是得参看源代码。

查到了cv::imwriteFlags（连接），发现了下表：

Enumerator
IMWRITE_JPEG_QUALITY	For JPEG, it can be a quality from 0 to 100 (the higher is the better). Default value is 95.
IMWRITE_JPEG_PROGRESSIVE	Enable JPEG features, 0 or 1, default is False.
IMWRITE_JPEG_OPTIMIZE	Enable JPEG features, 0 or 1, default is False.
IMWRITE_JPEG_RST_INTERVAL	JPEG restart interval, 0 - 65535, default is 0 - no restart.
IMWRITE_JPEG_LUMA_QUALITY	Separate luma quality level, 0 - 100, default is 0 - don't use.
IMWRITE_JPEG_CHROMA_QUALITY	Separate chroma quality level, 0 - 100, default is 0 - don't use.
IMWRITE_PNG_COMPRESSION	For PNG, it can be the compression level from 0 to 9. A higher value means a smaller size and longer compression time. Default value is 3.
IMWRITE_PNG_STRATEGY	One of cv::ImwritePNGFlags, default is IMWRITE_PNG_STRATEGY_DEFAULT.
IMWRITE_PNG_BILEVEL	Binary level PNG, 0 or 1, default is 0.
IMWRITE_PXM_BINARY	For PPM, PGM, or PBM, it can be a binary format flag, 0 or 1. Default value is 1.
IMWRITE_WEBP_QUALITY	For WEBP, it can be a quality from 1 to 100 (the higher is the better). By default (without any parameter) and for quality above 100 the lossless compression is used.

在保存PNG图像的时候，不只是有IMWRITE_PNG_COMPRESSION一个参数，还有其他好几个参数，其中IMWRITE_PNG_STRATEGY比较可以，进行了查看，发现其取值为：

Enumerator
IMWRITE_PNG_STRATEGY_DEFAULT
IMWRITE_PNG_STRATEGY_FILTERED
IMWRITE_PNG_STRATEGY_HUFFMAN_ONLY
IMWRITE_PNG_STRATEGY_RLE
IMWRITE_PNG_STRATEGY_FIXED

其中第三个IMWRITE_PNG_STRATEGY_HUFFMAN_ONLY感觉像是在说霍夫曼编码，估计是旋转不同的编码方式造成的吧，而且我保存的图像中大部分是0点，很有可能通过编码压缩了体积。

还需要验证，继续查找资料，通过bing搜索，前面的几个链接里面找到了下表（地址）：

IMWRITE_PNG_STRATEGY_DEFAULT	Use this value for normal data.
IMWRITE_PNG_STRATEGY_FILTERED	Use this value for data produced by a filter (or predictor).Filtered data consists mostly of small values with a somewhat random distribution. In this case, the compression algorithm is tuned to compress them better.
IMWRITE_PNG_STRATEGY_HUFFMAN_ONLY	Use this value to force Huffman encoding only (no string match).
IMWRITE_PNG_STRATEGY_RLE	Use this value to limit match distances to one (run-length encoding).
IMWRITE_PNG_STRATEGY_FIXED	Using this value prevents the use of dynamic Huffman codes, allowing for a simpler decoder for special applications.

进一步验证，使用如下代码保存无损PNG图像：

vector<int> compression_params;
compression_params.push_back(CV_IMWRITE_PNG_COMPRESSION);
compression_params.push_back(0);    // 无压缩png.
compression_params.push_back(IMWRITE_PNG_STRATEGY);
compression_params.push_back(IMWRITE_PNG_STRATEGY_DEFAULT);
imwrite(depthName, depthMat, compression_params);

输出的图像为128kB（131608字节），符合预期。

到这里应该没什么问题了，图像体积变小了是因为①图像中含有大量0点，②cv::imwrite保存图像是自动选择了某种编码方式压缩体积。