persistence.cpp:2197:ann.xml(1): There should be space between attributes in function icvXMLParseTag

62 阅读 0 评论 41 点赞

我是靠谱客的博主动人泥猴桃，最近开发中收集的这篇文章主要介绍persistence.cpp:2197:ann.xml(1): There should be space between attributes in function icvXMLParseTag，觉得挺不错的，现在分享给大家，希望可以做个参考。

概述

August 3, 2016
作者：dengshuai_super
出处：http://blog.csdn.net/dengshuai_super/article/details/52069634
声明：转载请注明作者及出处。

注意⚠️：此问题还不算解决，只是暂时用了别人的库，以后解决了，会回来记录。因为我是做笔记，所以建议大家直接看文章最后。
Opencv3.1.0 , Ubuntu16.04 ,开发板Microzed和Zedboard都试过了，出现同样问题。

charVec size:101
file /home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp, line 2197 terminate called after throwing an instance of 'cv::Exception'
what():/home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp:2197:error:(-212) /mnt/PRTest/resources/model/ann.xml(1):
There should be space between attributes in function icvXMLParseTag

http://www.th7.cn/Program/cp/201605/851893.shtml

/** @brief Loads algorithm from the file
@param filename Name of the file to read.
@param objname The optional name of the node to read (if empty, the first top-level node will be used)
This is static template method of Algorithm. It's usage is following (in the case of SVM):
@code
Ptr<SVM> svm = Algorithm::load<SVM>("my_svm_model.xml");
@endcode
In order to make this method work, the derived class must overwrite Algorithm::read(const
FileNode& fn).
*/
template<typename _Tp> static Ptr<_Tp> load(const String& filename, const String& objname=String())
{
FileStorage fs(filename, FileStorage::READ);
FileNode fn = objname.empty() ? fs.getFirstTopLevelNode() : fs[objname];
Ptr<_Tp> obj = _Tp::create();
obj->read(fn);

return !obj->empty() ? obj : Ptr<_Tp>();
}


/** @brief Reads algorithm parameters from a file storage
*/
virtual void read(const FileNode& fn) { (void)fn; }


/** @brief Reads algorithm from the file node
This is static template method of Algorithm. It's usage is following (in the case of SVM):
@code
Ptr<SVM> svm = Algorithm::read<SVM>(fn);
@endcode
In order to make this method work, the derived class must overwrite Algorithm::read(const
FileNode& fn) and also have static create() method without parameters
(or with all the optional parameters)
*/
template<typename _Tp> static Ptr<_Tp> read(const FileNode& fn)
{
Ptr<_Tp> obj = _Tp::create();
obj->read(fn);

return !obj->empty() ? obj : Ptr<_Tp>();
}

http://stackoverflow.com/questions/25831111/opencv-traincascade-build-intermediate-result/26301531#26301531

猜测一：xml本身出现了错误

Ubuntu 里面：/home/ds/桌面/EasyPR-master/resources/model/ann.xml

<?xml version="1.0"?>
<!opencv_storage>
<opencv_ml_ann_mlp>
<format>3</format>
<layer_sizes>120 40 65</layer_sizes>
<activation_function>SIGMOID_SYM</activation_function>
<f_param1>1.</f_param1>
<f_param2>1.</f_param2>

我故意把第二行加了一个”!”,Ubuntu里运行程序：出现：

OpenCV Error: Parsing error (resources/model/ann.xml(2):
file /home/ds/opencv-3.1.0/modules/core/src/persistence.cpp,
line 2249 terminate called after throwing an instance of 'cv::Exception'
what():
/home/ds/opencv-3.1.0/modules/core/src/persistence.cpp:2249:
error: (-212) resources/model/ann.xml(2):
<opencv_storage> tag is missing in function icvXMLParse

在已经做好opencv移植的arm linux里面运行，出现：

OpenCV Error: Parsing error (resources/model/ann.xml(1):
There should be space between attributes) in icvXMLParseTag,
file /home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp, line 2197 what():
/home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp:2197: error: (-212) resources/model/ann.xml(1):
There should be space between attributes in function icvXMLParseTag

然后我只把ann.xml 的第一行改为
<?xml version !="1.0"?>
在arm linux 报错和Ubuntu 上一样：

OpenCV Error: Parsing error (resources/model/ann.xml(1): Attribute name should be followed by '=') in icvXMLParseTag,
file /home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp,
line 2149 terminate called after throwing an instance of 'cv::Exception' what():
/home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp:2149:
error: (-212) resources/model/ann.xml(1): Attribute name should be followed by '=' in function icvXMLParseTag

我把ann.xml 的第一行改为
<?xml version = "1.0" >
arm linux显示错误：

OpenCV Error: Parsing error (resources/model/ann.xml(1):
There should be space between attributes) in icvXMLParseTag,
file /home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp,
line 2197 terminate called after throwing an instance of 'cv::Exception' what():
/home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp:2197:
error: (-212) resources/model/ann.xml(1):
There should be space between attributes in function icvXMLParseTag

结果arm上没检测出最右边？的问题，Ubuntu检测出来了。

于是我用opencv官方的例子人脸识别的那个（FaceDetection），在Ubuntu上可以顺利跑起来，移植到arm 上之后，出现了同样的问题：

WARNING: Could not load classifier cascade for nested objects OpenCV Error:
There should be space between attributes) in icvXMLParseTag,
file /home/ds/build/opencv3.1.0/modules/core/src/persistence.cpp, line 2197
terminate called after throwing an instance of 'cv::Exception'
what():
/home/ds/build/opencv-3.1.0/modules/core/src/persistence.cpp:2197: error: (-212) haarcascade_frontalface_alt.xml(1):
There should be space between attributes in function icvXMLParseTag
Aborted

说明这是通性的问题，正确的在Ubuntu上可以用的xml，移植到开发板上的arm linux里面时，就会出现
There should be space between attributes in function icvXMLParseTag 的问题。

猜测二：是arm上解析出现问题

开始时我怀疑是缺少libxml2库的问题导致了解析失败：（其实后来发现opencv是用自己的解析函数来解析xml）
于是我用ldd命令看Ubuntu程序链接了哪些库：
其中发现了xml2

 ldd demo_linux_amd64

打印出：

linux-vdso.so.1 =>
(0x00007ffdeb3b3000)
`libopencv_world.so.3.1 => /usr/local/lib/libopencv_world.so.3.1 (0x00007f62e0ffe000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007f62e0c61000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f62e0957000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007f62e0741000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f62e0378000)
libz.so.1 => /lib/x86_64-linux-gnu/libz.so.1 (0x00007f62e015d000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f62dff59000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f62dfd3c000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x00007f62dfb33000)
libpng12.so.0 => /lib/x86_64-linux-gnu/libpng12.so.0 (0x00007f62df90e000)`
libgtk-3.so.0 => /usr/lib/x86_64-linux-gnu/libgtk-3.so.0 (0x00007f62defdf000)
libgdk-3.so.0 => /usr/lib/x86_64-linux-gnu/libgdk-3.so.0 (0x00007f62ded03000)
libcairo.so.2 => /usr/lib/x86_64-linux-gnu/libcairo.so.2 (0x00007f62de9ef000)
libgdk_pixbuf-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libgdk_pixbuf-2.0.so.0 (0x00007f62de7cd000)
libgobject-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0 (0x00007f62de579000)
`libglib-2.0.so.0 => /lib/x86_64-linux-gnu/libglib-2.0.so.0 (0x00007f62de268000)`
libavcodec-ffmpeg.so.56 => /usr/lib/x86_64-linux-gnu/libavcodec-ffmpeg.so.56 (0x00007f62dce3e000)
libavformat-ffmpeg.so.56 => /usr/lib/x86_64-linux-gnu/libavformat-ffmpeg.so.56 (0x00007f62dca40000)
libavutil-ffmpeg.so.54 => /usr/lib/x86_64-linux-gnu/libavutil-ffmpeg.so.54 (0x00007f62dc7d1000)
libswscale-ffmpeg.so.3 => /usr/lib/x86_64-linux-gnu/libswscale-ffmpeg.so.3 (0x00007f62dc542000)
`/lib64/ld-linux-x86-64.so.2 (0x0000561e16755000)`
libgmodule-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libgmodule-2.0.so.0 (0x00007f62dc33d000)
libpangocairo-1.0.so.0 => /usr/lib/x86_64-linux-gnu/libpangocairo-1.0.so.0 (0x00007f62dc130000)
libX11.so.6 => /usr/lib/x86_64-linux-gnu/libX11.so.6 (0x00007f62dbdf5000)
libXi.so.6 => /usr/lib/x86_64-linux-gnu/libXi.so.6 (0x00007f62dbbe5000)
libXfixes.so.3 => /usr/lib/x86_64-linux-gnu/libXfixes.so.3 (0x00007f62db9df000)
libcairo-gobject.so.2 => /usr/lib/x86_64-linux-gnu/libcairo-gobject.so.2 (0x00007f62db7d5000)
libatk-1.0.so.0 => /usr/lib/x86_64-linux-gnu/libatk-1.0.so.0 (0x00007f62db5b0000)
libatk-bridge-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libatk-bridge-2.0.so.0 (0x00007f62db381000)
libepoxy.so.0 => /usr/lib/x86_64-linux-gnu/libepoxy.so.0 (0x00007f62db08b000)
libpangoft2-1.0.so.0 => /usr/lib/x86_64-linux-gnu/libpangoft2-1.0.so.0 (0x00007f62dae75000)
libpango-1.0.so.0 => /usr/lib/x86_64-linux-gnu/libpango-1.0.so.0 (0x00007f62dac29000)
libfontconfig.so.1 => /usr/lib/x86_64-linux-gnu/libfontconfig.so.1 (0x00007f62da9e5000)
libgio-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libgio-2.0.so.0 (0x00007f62da65d000)
libXinerama.so.1 => /usr/lib/x86_64-linux-gnu/libXinerama.so.1 (0x00007f62da459000)
libXrandr.so.2 => /usr/lib/x86_64-linux-gnu/libXrandr.so.2 (0x00007f62da24e000)
libXcursor.so.1 => /usr/lib/x86_64-linux-gnu/libXcursor.so.1 (0x00007f62da044000)
libXcomposite.so.1 => /usr/lib/x86_64-linux-gnu/libXcomposite.so.1 (0x00007f62d9e40000)
libXdamage.so.1 => /usr/lib/x86_64-linux-gnu/libXdamage.so.1 (0x00007f62d9c3d000)
libxkbcommon.so.0 => /usr/lib/x86_64-linux-gnu/libxkbcommon.so.0 (0x00007f62d99fe000)
libwayland-cursor.so.0 => /usr/lib/x86_64-linux-gnu/libwayland-cursor.so.0 (0x00007f62d97f5000)
libwayland-egl.so.1 => /usr/lib/x86_64-linux-gnu/libwayland-egl.so.1 (0x00007f62d95f3000)
libwayland-client.so.0 => /usr/lib/x86_64-linux-gnu/libwayland-client.so.0 (0x00007f62d93e5000)
libmirclient.so.9 => /usr/lib/x86_64-linux-gnu/libmirclient.so.9 (0x00007f62d9163000)
libXext.so.6 => /usr/lib/x86_64-linux-gnu/libXext.so.6 (0x00007f62d8f51000)
libpixman-1.so.0 => /usr/lib/x86_64-linux-gnu/libpixman-1.so.0 (0x00007f62d8ca8000)
libfreetype.so.6 => /usr/lib/x86_64-linux-gnu/libfreetype.so.6 (0x00007f62d89fe000)
libxcb-shm.so.0 => /usr/lib/x86_64-linux-gnu/libxcb-shm.so.0 (0x00007f62d87fa000)
libxcb-render.so.0 => /usr/lib/x86_64-linux-gnu/libxcb-render.so.0 (0x00007f62d85ef000)
libxcb.so.1 => /usr/lib/x86_64-linux-gnu/libxcb.so.1 (0x00007f62d83cd000)
libXrender.so.1 => /usr/lib/x86_64-linux-gnu/libXrender.so.1 (0x00007f62d81c3000)
libffi.so.6 => /usr/lib/x86_64-linux-gnu/libffi.so.6 (0x00007f62d7fba000)
libpcre.so.3 => /lib/x86_64-linux-gnu/libpcre.so.3 (0x00007f62d7d4a000)
libswresample-ffmpeg.so.1 => /usr/lib/x86_64-linux-gnu/libswresample-ffmpeg.so.1 (0x00007f62d7b2d000)
libva.so.1 => /usr/lib/x86_64-linux-gnu/libva.so.1 (0x00007f62d7911000)
libzvbi.so.0 => /usr/lib/x86_64-linux-gnu/libzvbi.so.0 (0x00007f62d7686000)
libxvidcore.so.4 => /usr/lib/x86_64-linux-gnu/libxvidcore.so.4 (0x00007f62d7372000)
libx265.so.79 => /usr/lib/x86_64-linux-gnu/libx265.so.79 (0x00007f62d6752000)
libx264.so.148 => /usr/lib/x86_64-linux-gnu/libx264.so.148 (0x00007f62d63ae000)
libwebp.so.5 => /usr/lib/x86_64-linux-gnu/libwebp.so.5 (0x00007f62d6152000)
libwavpack.so.1 => /usr/lib/x86_64-linux-gnu/libwavpack.so.1 (0x00007f62d5f28000)
libvpx.so.3 => /usr/lib/x86_64-linux-gnu/libvpx.so.3 (0x00007f62d5b04000)
libvorbisenc.so.2 => /usr/lib/x86_64-linux-gnu/libvorbisenc.so.2 (0x00007f62d585b000)
libvorbis.so.0 => /usr/lib/x86_64-linux-gnu/libvorbis.so.0 (0x00007f62d562e000)
libtwolame.so.0 => /usr/lib/x86_64-linux-gnu/libtwolame.so.0 (0x00007f62d540b000)
libtheoraenc.so.1 => /usr/lib/x86_64-linux-gnu/libtheoraenc.so.1 (0x00007f62d51cc000)
libtheoradec.so.1 => /usr/lib/x86_64-linux-gnu/libtheoradec.so.1 (0x00007f62d4fb1000)
libspeex.so.1 => /usr/lib/x86_64-linux-gnu/libspeex.so.1 (0x00007f62d4d98000)
libsnappy.so.1 => /usr/lib/x86_64-linux-gnu/libsnappy.so.1 (0x00007f62d4b90000)
libshine.so.3 => /usr/lib/x86_64-linux-gnu/libshine.so.3 (0x00007f62d4982000)
libschroedinger-1.0.so.0 => /usr/lib/x86_64-linux-gnu/libschroedinger-1.0.so.0 (0x00007f62d46ad000)
libopus.so.0 => /usr/lib/x86_64-linux-gnu/libopus.so.0 (0x00007f62d4463000)
`libopenjpeg.so.5 => /usr/lib/x86_64-linux-gnu/libopenjpeg.so.5 (0x00007f62d423f000)`
libmp3lame.so.0 => /usr/lib/x86_64-linux-gnu/libmp3lame.so.0 (0x00007f62d3fca000)
libgsm.so.1 => /usr/lib/x86_64-linux-gnu/libgsm.so.1 (0x00007f62d3dbc000)
libcrystalhd.so.3 => /usr/lib/x86_64-linux-gnu/libcrystalhd.so.3 (0x00007f62d3ba0000)
liblzma.so.5 => /lib/x86_64-linux-gnu/liblzma.so.5 (0x00007f62d397e000)
libssh-gcrypt.so.4 => /usr/lib/x86_64-linux-gnu/libssh-gcrypt.so.4 (0x00007f62d3734000)
librtmp.so.1 => /usr/lib/x86_64-linux-gnu/librtmp.so.1 (0x00007f62d3518000)
libmodplug.so.1 => /usr/lib/x86_64-linux-gnu/libmodplug.so.1 (0x00007f62d318d000)
libgme.so.0 => /usr/lib/x86_64-linux-gnu/libgme.so.0 (0x00007f62d2f3e000)
libbluray.so.1 => /usr/lib/x86_64-linux-gnu/libbluray.so.1 (0x00007f62d2cf5000)
libgnutls.so.30 => /usr/lib/x86_64-linux-gnu/libgnutls.so.30 (0x00007f62d29c5000)
libbz2.so.1.0 => /lib/x86_64-linux-gnu/libbz2.so.1.0 (0x00007f62d27b4000)
libatspi.so.0 => /usr/lib/x86_64-linux-gnu/libatspi.so.0 (0x00007f62d2585000)
libdbus-1.so.3 => /lib/x86_64-linux-gnu/libdbus-1.so.3 (0x00007f62d2338000)
libharfbuzz.so.0 => /usr/lib/x86_64-linux-gnu/libharfbuzz.so.0 (0x00007f62d20da000)
libthai.so.0 => /usr/lib/x86_64-linux-gnu/libthai.so.0 (0x00007f62d1ed0000)
libexpat.so.1 => /lib/x86_64-linux-gnu/libexpat.so.1 (0x00007f62d1ca7000)
libselinux.so.1 => /lib/x86_64-linux-gnu/libselinux.so.1 (0x00007f62d1a85000)
libresolv.so.2 => /lib/x86_64-linux-gnu/libresolv.so.2 (0x00007f62d186a000)
libmircommon.so.5 => /usr/lib/x86_64-linux-gnu/libmircommon.so.5 (0x00007f62d1638000)
libmirprotobuf.so.3 => /usr/lib/x86_64-linux-gnu/libmirprotobuf.so.3 (0x00007f62d13d7000)
libboost_system.so.1.58.0 => /usr/lib/x86_64-linux-gnu/libboost_system.so.1.58.0 (0x00007f62d11d2000)
libprotobuf-lite.so.9 => /usr/lib/x86_64-linux-gnu/libprotobuf-lite.so.9 (0x00007f62d0fa1000)
libXau.so.6 => /usr/lib/x86_64-linux-gnu/libXau.so.6 (0x00007f62d0d9c000)
libXdmcp.so.6 => /usr/lib/x86_64-linux-gnu/libXdmcp.so.6 (0x00007f62d0b96000)
libsoxr.so.0 => /usr/lib/x86_64-linux-gnu/libsoxr.so.0 (0x00007f62d0931000)
libnuma.so.1 => /usr/lib/x86_64-linux-gnu/libnuma.so.1 (0x00007f62d0725000)
libogg.so.0 => /usr/lib/x86_64-linux-gnu/libogg.so.0 (0x00007f62d051c000)
liborc-0.4.so.0 => /usr/lib/x86_64-linux-gnu/liborc-0.4.so.0 (0x00007f62d029b000)
libgcrypt.so.20 => /lib/x86_64-linux-gnu/libgcrypt.so.20 (0x00007f62cffba000)
libgssapi_krb5.so.2 => /usr/lib/x86_64-linux-gnu/libgssapi_krb5.so.2 (0x00007f62cfd6f000)
libhogweed.so.4 => /usr/lib/x86_64-linux-gnu/libhogweed.so.4 (0x00007f62cfb3c000)
libnettle.so.6 => /usr/lib/x86_64-linux-gnu/libnettle.so.6 (0x00007f62cf906000)
libgmp.so.10 => /usr/lib/x86_64-linux-gnu/libgmp.so.10 (0x00007f62cf685000)
`libxml2.so.2 => /usr/local/lib/libxml2.so.2 (0x00007f62cf31e000)`
libp11-kit.so.0 => /usr/lib/x86_64-linux-gnu/libp11-kit.so.0 (0x00007f62cf0b8000)
libidn.so.11 => /usr/lib/x86_64-linux-gnu/libidn.so.11 (0x00007f62cee85000)
libtasn1.so.6 => /usr/lib/x86_64-linux-gnu/libtasn1.so.6 (0x00007f62cec72000)
libsystemd.so.0 => /lib/x86_64-linux-gnu/libsystemd.so.0 (0x00007f62cebec000)
libgraphite2.so.3 => /usr/lib/x86_64-linux-gnu/libgraphite2.so.3 (0x00007f62ce9c7000)
libdatrie.so.1 => /usr/lib/x86_64-linux-gnu/libdatrie.so.1 (0x00007f62ce7bf000)
libboost_filesystem.so.1.58.0 => /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.58.0 (0x00007f62ce5a6000)
libgomp.so.1 => /usr/lib/x86_64-linux-gnu/libgomp.so.1 (0x00007f62ce384000)
libgpg-error.so.0 => /lib/x86_64-linux-gnu/libgpg-error.so.0 (0x00007f62ce16f000)
libkrb5.so.3 => /usr/lib/x86_64-linux-gnu/libkrb5.so.3 (0x00007f62cde9d000)
libk5crypto.so.3 => /usr/lib/x86_64-linux-gnu/libk5crypto.so.3 (0x00007f62cdc6e000)
libcom_err.so.2 => /lib/x86_64-linux-gnu/libcom_err.so.2 (0x00007f62cda69000)
libkrb5support.so.0 => /usr/lib/x86_64-linux-gnu/libkrb5support.so.0 (0x00007f62cd85e000)
libkeyutils.so.1 => /lib/x86_64-linux-gnu/libkeyutils.so.1 (0x00007f62cd659000)

可以看出在Ubuntu跑这个程序需要这么多动态链接库的支持，我们在开发板上有好多是用不到的，其中一部分在交叉编译安装opencv时已经禁用掉了。
我把我认为比较重要的库都标注成了绿色。（只是我个人理解，水平有限。）
文章：http://ebaina.com/bbs/forum.php?mod=viewthread&tid=7828&page=1&extra=#pid16931 上面说：

解决方法：
1.对于grfmt_png.cpp的各种undefined reference to，原因是交叉编译环境缺少libpng，解决办法是下载libpng，然后进行交叉编译，编译后生成include和lib文件夹，分别把include和lib下的文件拷贝到交叉编译环境的include和lib文件下，我的也就是上面的/usr/local/hi-arm/arm-hisiv100-linux/arm-hisiv100-linux-uclibcgnueabi/include/和/usr/local/hi-arm/arm-hisiv100-linux/arm-hisiv100-linux-uclibcgnueabi/lib/
然后再进行编译，记得编译命令上加上-lpng，然后发现grfmt_png.cpp的错误都没有了。
2.对于grfmt_jpeg.cpp的各种undefined reference to，是交叉编译环境缺少了libjpeg，解决办法参考上面的1，下载libjpeg，交叉编译，拷贝include和lib文件到交叉编译环境的include和lib下，编译加上-ljpeg
3对于persistence.cpp的各种undefined reference to，是交叉编译环境缺少了libxml2，解决办法同上，编译时加上-xlm2后，这个错误就没了。

受到这个启发，我决定试试。
于是我打算重新交叉编译安装opencv3.1.0和支持xml，jpeg，png的库。
编译libxml2参考：
http://blog.csdn.net/lanyou1900/article/details/47782067

安装libiconv:
http://ftp.gnu.org/pub/gnu/libiconv/libiconv-1.13.1.tar.gz

./configure CC=arm-none-linux-gnueabi-gcc --host=arm-none-linux-gnueabi --prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv //(这个文件夹自己建立的)
make
make install

安装libz：

./configure --prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz
`修改makefile：将30行左右改为：（将LDSHARED 改为交叉编译器）
LDSHARED=arm-none-linux-gnueabi-gcc -shared -Wl,-soname,libz.so.1,--version-script,zlib.map`
`要不然会出现错误：/usr/bin/ld: adler32.lo：普通ELF重定位（M: 40）
/usr/bin/ld: adler32.lo：普通ELF重定位（M: 40）
adler32.lo: 无法添加符号: 文件格式错误
collect2: error: ld returned 1 exit status
Makefile:163: recipe for target 'libz.so.1.2.8' failed`
make CC=arm-none-linux-gnueabi-gcc LD=arm-none-linux-gnueabi-ld AR=arm-none-linux-gnueabi-ar
make install

安装libxml2：

./configure --host=arm-none-linux-gnueabi CC=arm-none-linux-gnueabi-gcc AR=arm-none-linux-gnueabi-ar LD=arm-none-linux-gnueabi-ld ANLIB=arm-none-linux-gnueabi-ranlib --prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libxml2 --with-zlib=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz
--with-iconv=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv
make
`出现错误：parserInternals.c:36:18: fatal error: zlib.h: No such file or directory
compilation terminated.
Makefile:1248: recipe for target 'parserInternals.lo' failed
make[2]: *** [parserInternals.lo] Error 1
make[2]: Leaving directory '/home/ds/build/xml/libxml2-2.9.4'
Makefile:1440: recipe for target 'all-recursive' failed
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory '/home/ds/build/xml/libxml2-2.9.4'
Makefile:859: recipe for target 'all' failed
make: *** [all] Error 2
于是我去makefile文件，搜索libiconv，在对应的位置把libz的动态链接库和位置和头文件都添加上
XML_INCLUDEDIR = -I${includedir}/libxml2 -I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv/include -I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/include
XML_LIBS = -lxml2 -lz
-L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv/lib -liconv
-lm -L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/lib -lz
CPPFLAGS =
-I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv/include -I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/include
ICONV_LIBS = -L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv/lib -liconv -L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/lib -lz
这个错误解决。又出现下面python错误，不过网上说没关系，我暂时忽略了。我看了以下makefile文件，应该是编译时python指定的路径不对，或者缺少库。如果不成功再回来解决。
`
make install

出现错误：

cc1: warning: include location "/usr/include/python2.7" is unsafe for cross-compilation [-Wpoison-system-directories]
In file included from /usr/include/python2.7/Python.h:8:0,
from libxml.c:14:
/usr/include/python2.7/pyconfig.h:30:52: fatal error: arm-linux-gnueabi/python2.7/pyconfig.h: No such file or directory
compilation terminated.
Makefile:623: recipe for target 'libxml.lo' failed
make[4]: *** [libxml.lo] Error 1
make[4]: Leaving directory '/home/ds/build/xml/libxml2-2.9.4/python'
Makefile:684: recipe for target 'install-recursive' failed
make[3]: *** [install-recursive] Error 1
make[3]: Leaving directory '/home/ds/build/xml/libxml2-2.9.4/python'
Makefile:841: recipe for target 'install' failed
make[2]: *** [install] Error 2
make[2]: Leaving directory '/home/ds/build/xml/libxml2-2.9.4/python'
Makefile:1440: recipe for target 'install-recursive' failed
make[1]: *** [install-recursive] Error 1
make[1]: Leaving directory '/home/ds/build/xml/libxml2-2.9.4'
Makefile:1743: recipe for target 'install' failed
make: *** [install] Error 2

网上说是python错误无视即可。http://www.aichengxu.com/view/40322
用链接里例子测试了下，成功生成了可执行文件（不过我没移植到板子上测试，等全部的库弄完了再单独测试这个）。

#include <stdio.h>
#include <libxml/parser.h>
#include <libxml/tree.h>
int main(int argc, char **argv)
{
xmlDocPtr doc = NULL;
xmlNodePtr root_node = NULL, node = NULL, node1 = NULL;
doc = xmlNewDoc(BAD_CAST "1.0");
root_node = xmlNewNode(NULL, BAD_CAST "root");
xmlDocSetRootElement(doc, root_node);
xmlNewChild(root_node, NULL, BAD_CAST "node1",BAD_CAST "content of node1");
node=xmlNewChild(root_node, NULL, BAD_CAST "node3",BAD_CAST"node has attributes");
xmlNewProp(node, BAD_CAST "attribute", BAD_CAST "yes");
node = xmlNewNode(NULL, BAD_CAST "node4");
node1 = xmlNewText(BAD_CAST"other way to create content");
xmlAddChild(node, node1);
xmlAddChild(root_node, node);
xmlSaveFormatFileEnc(argc > 1 ? argv[1] : "-", doc, "UTF-8", 1);
xmlFreeDoc(doc);
xmlCleanupParser();
xmlMemoryDump();
return 0;
}

arm-none-linux-gnueabi-gcc -I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libxml2/include/libxml2 -I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/include -I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv/include
-L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libxml2/lib -L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/lib -L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libiconv/lib -lz -liconv -lxml2 xml2test.c -o test
//因为libxml2依赖那两个库，因此要加上，否则出错。

接下来安装：libjpeg
http://www.ijg.org/

./configure --host=arm-none-linux-gnueabi CC=arm-none-linux-gnueabi-gcc
--prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libjpeg
make
make install

安装libpng：
https://sourceforge.net/projects/libpng/files/
或者http://download.csdn.net/download/damys/9326857

./configure --host=arm-none-linux-gnueabi CC=arm-none-linux-gnueabi-gcc
--prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libpng --with-zlib=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/

出错：

checking for zlibVersion in -lz... no
checking for /home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/zlibVersion in -lz... no
configure: error: zlib not installed

于是加上zlib库具体的位置

./configure --host=arm-none-linux-gnueabi CC=arm-none-linux-gnueabi-gcc
--prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libpng LIBS=-L/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/lib CPPFLAGS=-I/home/ds/arm-2013.05/x86_64-linux-gnu/usr/libz/include
make
make install

成功。

编译时可以加上以下参数：

如果你要编译一个库的源代码，可以把它编译成静态库，也可以把它编译成动态库。如果你想编译成动态库，就用 --enable-shared参数；如果你想编译成静态库，就用--enable-static参数。

交叉编译 x264,xvid,ffmpeg
为了使OpenCV能处理视频，我们要事先交叉编译ffmpeg,而ffmpeg又是依赖x264和xvid的。
安装yasm（x264需要用到的汇编编译器）：
http://yasm.tortall.net/Download.html

./configure --host=arm-none-linux-gnueabi CC=arm-none-linux-gnueabi-gcc
--prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/yasm
make

交叉编译x264

CC=arm-none-linux-gnueabi-gcc ./configure --host=arm-none-linux-gnueabi --disable-asm
--prefix=/home/ds/arm-2013.05/x86_64-linux-gnu/usr/x264
make
make install

交叉编译libxvid:

猜测三：xml编码问题

用UltraEdit把编码都改为UNIX utf-8 ，发现还是出现相同错误。（后来发现从ubuntu拷贝到板子上本身编码是没问题的。）
查看opencv3.1源码，persistance.cpp —>

static void
icvXMLParse( CvFileStorage* fs )
{
char* ptr = fs->buffer_start;
CvStringHashNode *key = 0, *key2 = 0;
CvAttrList* list = 0;
int tag_type = 0;
// CV_XML_INSIDE_TAG is used to prohibit leading comments
ptr = icvXMLSkipSpaces( fs, ptr, CV_XML_INSIDE_TAG );
if( memcmp( ptr, "<?xml", 5 ) != 0 ) //比较前5个字节，因为编码格式是8位。每个字符占8位。
CV_PARSE_ERROR( "Valid XML should start with '<?xml ...?>'" );
ptr = icvXMLParseTag( fs, ptr, &key, &list, &tag_type );
/*{
const char* version = cvAttrValue( list, "version" );
if( version && strncmp( version, "1.", 2 ) != 0 )
CV_Error( CV_StsParseError, "Unsupported version of XML" );
}*/
// we support any 8-bit encoding, so we do not need to check the actual encoding.//支持任何8位的编码
// we do not support utf-16, but in the case of utf-16 we will not get here anyway.
/*{
const char* encoding = cvAttrValue( list, "encoding" );
if( encoding && strcmp( encoding, "ASCII" ) != 0 &&
strcmp( encoding, "UTF-8" ) != 0 &&
strcmp( encoding, "utf-8" ) != 0 )
CV_PARSE_ERROR( "Unsupported encoding" );
}*/
while( *ptr != '' )
{
ptr = icvXMLSkipSpaces( fs, ptr, 0 );
if( *ptr != '' )
{
CvFileNode* root_node;
ptr = icvXMLParseTag( fs, ptr, &key, &list, &tag_type );
if( tag_type != CV_XML_OPENING_TAG ||
strcmp(key->str.ptr,"opencv_storage") != 0 )
CV_PARSE_ERROR( "<opencv_storage> tag is missing" );
root_node = (CvFileNode*)cvSeqPush( fs->roots, 0 );
ptr = icvXMLParseValue( fs, ptr, root_node, CV_NODE_NONE );
ptr = icvXMLParseTag( fs, ptr, &key2, &list, &tag_type );
if( tag_type != CV_XML_CLOSING_TAG || key != key2 )
CV_PARSE_ERROR( "</opencv_storage> tag is missing" );
ptr = icvXMLSkipSpaces( fs, ptr, 0 );
}
}
assert( fs->dummy_eof != 0 );
}

/** @overload
@param source Name of the file to open or the text string to read the data from. Extension of the
file (.xml or .yml/.yaml) determines its format (XML or YAML respectively). Also you can append .gz
to work with compressed files, for example myHugeMatrix.xml.gz. If both FileStorage::WRITE and
FileStorage::MEMORY flags are specified, source is used just to specify the output file format (e.g.
mydata.xml, .yml etc.).
@param flags Mode of operation. See
FileStorage::Mode
@param encoding Encoding of the file. Note that UTF-16 XML encoding is not supported currently and
you should use 8-bit encoding instead of it.
*/
CV_WRAP FileStorage(const String& source, int flags, const String& encoding=String());
/** @overload */
FileStorage(CvFileStorage* fs, bool owning=true);

可以看出任何8位的字符编码，opencv都是支持的。因此排除字符编码因素。

猜测四：是xml文件过大出现问题

在网上找简单的xml读写例子：

//readxml.cpp
#include <stdio.h>

#include <iostream>

#include "opencv2/opencv.hpp"

using namespace std;
using namespace cv;
int main(int argc, char* argv[])
{
//创建XML文件存储

CvFileStorage* fs = cvOpenFileStorage("jarvischu.xml",
0,
//用于存储的内存，如果为NULL,则会使用一个临时内存

CV_STORAGE_WRITE,
"utf-8"//编码

);
//写入数据

cvWriteInt(fs,"frame_count",10);
//--写Int

cvWriteReal(fs,"pi",3.14);
//--写Float

//--写结构体

cvStartWriteStruct(fs,"frame_size",CV_NODE_MAP,"id_size" );// 有名称（会新建一个标签）

cvWriteInt(fs,"width",320);
cvWriteInt(fs,"height",200);
cvEndWriteStruct(fs);
cvStartWriteStruct(fs,"author_info",CV_NODE_SEQ,"id_author");// 无名称

cvWriteString(fs,0,"JarvisChu");
cvWriteString(fs,0,"China");
cvEndWriteStruct(fs);
//--写矩阵

unsigned char vec[]={1,2,3,4,5,6,7,8,9};
CvMat mat = cvMat(3,3,CV_8UC1,vec);
cvWrite(fs,"Matrix",&mat);
//--写注释

cvWriteComment(fs,"This is a example for operatoring the xml file",0);//不为0表示放在当前行，0表示新行

//释放

cvReleaseFileStorage(&fs);
//打开XML文件

fs = cvOpenFileStorage("jarvischu.xml",
0,
//用于存储的内存，如果为NULL,则会使用一个临时内存

CV_STORAGE_READ,
"GB2312"//编码
GB2312
);
//读取数据

int f = cvReadIntByName(fs,0,"frame_count",0);//读取有名字的Int

//--读取元素数据

CvFileNode* fn = cvGetFileNodeByName(fs,0,"frame_size"); //先读取父元素（标签，节点）

int width = cvReadIntByName(fs,fn,"width"); //通过父元素读取子元素

int height = cvReadIntByName(fs,fn,"height");
//--读取元素内的顺序数据

fn = cvGetFileNodeByName(fs,0,"author_info");//先读取元素

CvSeq* seq = fn->data.seq;
//获取元素的顺序流

const char* name = cvReadString((CvFileNode*)cvGetSeqElem(seq,0));//读取顺序流中数据 0

const char* country = cvReadString((CvFileNode*)cvGetSeqElem(seq,1));//读取顺序流中数据 1

cout<<"frame_count:"<<f<<endl<<"width:"<<width<<endl<<"height"<<height<<endl;
cout <<"name:"<<name<<endl<<"country:"<<country<<endl;
return 0;
}

和读写yml的例子：

readyml.cpp
#include <stdio.h>
#include "opencv2/opencv.hpp"
#include "opencv2/ml.hpp"
#include <iostream>
#include <time.h>
using namespace cv;
using namespace std;
int main(int argc, char **argv)
{
/*
FileStorage fs("test.yml", FileStorage::WRITE);
fs << "frameCount" << 5;
time_t rawtime; time(&rawtime);
fs << "calibrationDate" << asctime(localtime(&rawtime));
Mat cameraMatrix = (Mat_<double>(3,3) << 1000, 0, 320, 0, 1000, 240, 0, 0, 1); //又一种Mat初始化方式
Mat distCoeffs = (Mat_<double>(5,1) << 0.1, 0.01, -0.001, 0, 0);
fs << "cameraMatrix" << cameraMatrix << "distCoeffs" << distCoeffs;
//features为一个大小为3的向量,其中每个元素由随机数x,y和大小为8的uchar数组组成
fs << "features" << "[";
for( int i = 0; i < 3; i++ )
{
int x = rand() % 640;
int y = rand() % 480;
uchar lbp = rand() % 256;
fs << "{:" << "x" << x << "y" << y << "lbp" << "[:";
for( int j = 0; j < 8; j++ )
fs << ((lbp >> j) & 1);
fs << "]" << "}";
}
fs << "]";
fs.release();*/
FileStorage fs("test.yml", FileStorage::READ);
//方式一: []操作符
int frameCount = (int)fs["frameCount"];
//方式二: FileNode::operator >>()
string date;
fs["calibrationDate"] >> date;
Mat cameraMatrix2, distCoeffs2;
fs["cameraMatrix"] >> cameraMatrix2;
fs["distCoeffs"] >> distCoeffs2;
//注意FileNodeIterator的使用,似乎只能用一维数组去读取里面所有的数据
FileNode features = fs["features"];
FileNodeIterator it = features.begin(), it_end = features.end();
int idx = 0;
std::vector<uchar> lbpval;
for( ; it != it_end; ++it, idx++ )
{
cout << "feature #" << idx << ": ";
cout << "x=" << (int)(*it)["x"] << ", y=" << (int)(*it)["y"] << ", lbp: (";
(*it)["lbp"] >> lbpval;
//直接读出一维向量
for( int i = 0; i < (int)lbpval.size(); i++ )
cout << " " << (int)lbpval[i];
cout << ")" << endl;
}
fs.release();
return 0;
}

重点内容结果：在arm上写xml并保存成功，但是读取时出现相同错误。排除这一猜测。
但是在arm 上读写保存yml文件都成功。

猜测五：xml路径或者opencv中的load加载问题


FILE* f=fopen(kDefaultAnnPath, "rb+");//http://baike.baidu.com/link?url=1AGB1MzmqkrH0iMjhCb2KTxGIp8u5rb7zebfMGl7gKma5uIcbEUwIsxvqyuo27cA1kzO6yYbKhl-dxNV7ZuBYa
fseek(f, 0, SEEK_END);//函数设置文件指针stream的位置。如果执行成功，stream将指向以fromwhere（偏移起始位置：文件头0(SEEK_SET)，当前位置1(SEEK_CUR)，文件尾2(SEEK_END)）为基准，偏移offset（指针偏移量）个字节的位置。如果执行失败(比如offset超过文件自身大小)，则不改变stream指向的位置。（让stream指针指向文件尾）
long l = ftell(f);//函数 ftell 用于得到文件位置指针当前位置相对于文件首的偏移字节数。在随机方式存取文件时，由于文件位置频繁的前后移动，程序不容易确定文件的当前位置。(为了得到文件的字节数)
fseek(f, 0, SEEK_SET);//让stream指针指向文件头，int fseek(FILE *stream, long offset, int fromwhere);
char * str =new char[1024*1024];
fread(str, l, 1, f);//函数原型size_t fread ( void *buffer, size_t size, size_t count, FILE *stream) ;参 数: buffer用于接收数据的内存地址;size : 要读的每个数据项的字节数，单位是字节; count: 要读count个数据项，每个数据项size个字节；stream：输入流
fclose(f);
str[l] = '';//把str最后一个字符设置成,字符串结束符
ann_=ml::ANN_MLP::loadFromString<ml::ANN_MLP>(str);

还是报相同错误。排除。

猜测六：是opencv版本问题

根据以上的结果基本可以断定原因是：在opencv读xml文件的方式在PC（Ubuntu）上和在arm 中的linux上是不同的。
因此我要交叉编译opencv2.4.9，并且用他的库编译一个读写xml文件的程序，移植到板子上是否有问题。
用上面那个readxml.cpp的例子，用opencv-2.4.9的库交叉编译出来放到板子上跑，还是写xml和保存没问题但是读的时候还是相同的错误。排除。

猜测七：是交叉编译器的问题

开发板从SD卡启动，我查看内核版本：

ds@ds-Lenovo:~/build/filesys_zedboard/备份$ file uImage
uImage: u-boot legacy uImage, Linux-4.0.0-xilinx, Linux/ARM, OS Kernel Image (Not compressed), 3562248 bytes, Mon Mar 28 05:19:08 2016, Load Address: 0x00008000, Entry Point: 0x00008000, Header CRC: 0x94BFB3BC, Data CRC: 0x72900298

我之前安装交叉编译器之前未安装32位库，不知道这个有没有影响。
重新安装交叉编译器，之前用的是从网上下载的arm-2013.05-24-arm-none-linux-gnueabi-x86_64-linux-gnu.tar.bz2。
现在用xilinx-2011.09-50-arm-xilinx-linux-gnueabi.bin,进行交叉编译readxml.cpp
。移植到zedboard上，还是相同的错误。（也有可能是交叉编译器版本太低，以后用高版本的试一下）

猜测八：是交叉编译opencv时出问题了。

于是打算重新交叉编译opencv3.1.0。
我按照
http://www.eefocus.com/beiyangguangdian/blog/14-03/302119_52e49.html 的方法移植的，可是老是出现交叉编译器include中c++头文件math.h或者cmath的问题。找了网上说什么的都有，有说是交叉编译器版本应该用新版本，有的说是ffmpeg版本的问题。
我分别用opencv3.1版本和opencv2.4.9版本都试了，如果只加上with_jpeg,png,tiff,v4l不出错，一旦加上就出上边math.h或者cmath里面的错误❌。暂时没有解决。
不过xml这个问题，我觉得应该是交叉编译器或者opencv版本或者编译时依赖库的问题（不知道和ubuntu是64位的有没有关系），以后这个问题解决了，我再回过头来记录一下。

现在阶段我用了XAPP1167里面的demo，里面有现成的opencv_install/arm_linux ,有编译好的opencv库，而且支持ffmpeg。它的opencv是2.4.5版本的。
我用它提供的库交叉编译了opencv自带的人脸识别程序和自己的xml读写程序都成功了。（交叉编译器是arm-2014.05-28-arm-none-eabi-i686版本的，其实我觉得和交叉编译器没关系，这里只是记录一下）