OpenCV移植

OpenCV是目前开源项目中最著名的基于机器视觉方向的图像处理的开发包，目前已经有被移植到嵌入式Linux环境上。本文介绍了OpenCV交叉编译的基本步骤。

一、编译环境及库文件

linux环境：fedora14

交叉编译：4.5.1
arm板子：tiny6410

libz:        zlib-1.2.7
libjpeg:       jpegsrc.v7
libpng:       libpng-1.5.10
libyasm:       yasm-1.2.0
opencv:        opencv-2.4.2
libx264:       x264-snapshot-20120608-2245
libxvid:       xvidcore-1.3.2
lffmpeg:       ffmpeg-0.10.3

cmake:         cmake-2.8.8-Linux-i386.tar,gz

二、opencv-2.4.1的移植过程

1、libz的交叉编译：

# ./configure --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi --shared

修改makefile

CC=arm-linux-gcc

AR=arm-linux-ar rc

RANLIB=arm-linux-ranlib

CDSHARED=arm-linux-gcc -shared -wl, -soname,libz.so.1,--version-script,zl

lib.map
#make 
#make install

2、libjpeg的交叉编译：

#./configure --host=arm-linux --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi --enable-shared --enable-static
#make install


3、libpng的交叉编译

#./configure --host=arm-linux --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi --enable-shared --enable-static 
#make
#make install

4、yasm的交叉编译:

#./configure --host=arm-linux --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi --enable-shared --enable-static 
#make
#make install


5、libx264的交叉编译:

#CC=arm-linux-gcc ./configure --enable-shared --host=arm-linux --disable-asm --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi
#make
#make install
6、libxvid的交叉编译:

#cd build/generic
#./configure --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi --host=arm-linux  --disable-assembly
#make
#make install

7、ffmpeg的交叉编译:

#./configure --prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi --enable-shared --disable-static --enable-gpl --enable-cross-compile --arch=arm --disable-stripping --target-os=linux --enable-libx264 --enable-libxvid --cc=arm-linux-gcc --enable-swscale
#make
#make install

8.cmake-gui安装

到官网上下载已经编译好的cmake-2.8.8-Linux-i386.tar,gz，解压就可以了

 

9、opencv2.4.1的交叉编译:

用到的主要目录说明：
交叉编译工具链所在目录     /opt/FriendlyARM/toolschain/4.5.1/

安装opencv的目录         /usr/local/

opencv源码所在目录          /home/6410/opencv/opencv-2.4.2/

编译好的opencv库所在目录      /home/6410/opencv/opencv-2.4.2/build

#mkdir build
#cd build
#cmake-gui
选择源代码目录/home/6410/opencv/opencv-2.4.2/

选择Build目录：/home/6410/opencv/opencv-2.4.2/build
点击Configure，保持generator为Unix Makefiles，选择Specify options for cross-compiling，点击Next,
Operating System填写arm-linux
C Compilers填写opt/FriendlyARM/toolschain/4.5.1/bin/arm-linux-gcc
C++ Compilers填写opt/FriendlyARM/toolschain/4.5.1//bin/arm-linux-g++
程序库的Target Root填写opt/FriendlyARM/toolschain/4.5.1/

然后点击Finish。
默认安装目录为/usr/local，点击Generate生成Makefile。

#make


错误一：

Linking CXX executable ../../bin/opencv_createsamples
../../lib/libopencv_core.so: undefined reference to `clock_gettime'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFerrorHandler'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFrealloc'
../../lib/libopencv_core.so: undefined reference to `pthread_key_create

解决方法：
修改CMakeCache.txt，CMAKE_EXE_LINKER_FLAGS原来为空，加上-lpthread -lrt，重新编译，错误消除

错误二：
Linking CXX executable ../../bin/opencv_createsamples
../../lib/libopencv_highgui.so: undefined reference to `_TIFFerrorHandler'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFrealloc'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFmalloc'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFmemcpy'
../../lib/libopencv_highgui.so: undefined reference to `TIFFOpen'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFfree'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFwarningHandler'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFmemcmp'
../../lib/libopencv_highgui.so: undefined reference to `_TIFFmemset'
    解决方法：修改CMakeCache.txt，WITH_TIFF:BOOL=ON，改成OFF，重新编译，错误消除。

#make install
三、测试

测试程序：
#include <cv.h>  
#include <cxcore.h>  
#include <highgui.h>  
      
      
int main(int argc,char **argv)  
{  
    CvCapture* capture = NULL;  
    IplImage* frame = NULL;  
      
    if(!(capture = cvCaptureFromCAM(-1)))  
    {  
        fprintf(stderr, "Can not open camera./n");  
        return -1;  
    }  
      
    cvNamedWindow("video", 1);  
      
    while(frame = cvQueryFrame( capture ) )  
    {  
        cvShowImage("video", frame);  
    }  
      
    cvDestroyWindow("video");  
    cvReleaseCapture(&capture);  

    return 0;  
}  

编译：
＃arm-linux-g++ -o opencv_test opencv_test.cpp -I /usr/local/include -L /usr/local/lib -lopencv_core -lopencv_highgui -lpthread -lrt