opencv应用程序移植到hi3559板卡

opencv应用程序移植到hi3559板卡

opencv在hi3559平台的移植网络上有很多文章，此处不做赘述。很多新手将opencv使用交叉编译工具编译好后，不知道怎么通过交叉编译将opencv应用程序正常编译，且在板卡上运行，本文章就是为新手做个指引，高手可以忽略。

应用程序

本人在hi3559上移植了opencv3.4.1版本；此处以柱面投影算法作为例子：
#include <opencv2/core/core.hpp>
#include <opencv2/imgcodecs/imgcodecs.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <math.h>

// remapping an image by creating cylind effects
void cylind_projection(const cv::Mat &image, cv::Mat &result) {

float angle = CV_PI/4;
float f = image.rows / (tan(angle/2) * 2);

// the map functions
cv::Mat srcX(image.rows,image.cols,CV_32F); // x-map
cv::Mat srcY(image.rows,image.cols,CV_32F); // y-map

// creating the mapping
for (int i=0; i<image.rows; i++) {
    for (int j=0; j<image.cols; j++) {

        srcX.at<float>(i,j)= f * tan(j/f - angle/2) + image.cols/2;
        srcY.at<float>(i,j)= (i - image.rows/2) / cos(j/f - angle/2) + image.rows/2;
    }
}

// applying the mapping
cv::remap(image,  // source image
          result, // destination image
          srcX,   // x map
          srcY,   // y map
          cv::INTER_LINEAR); // interpolation method

}

int main()
{
// open image
cv::Mat image= cv::imread("./boldt.jpg",0);
if(image.data == NULL)
{
printf(“imread failed\n”);
return -1;
}

// remap image
cv::Mat result;
cylind_projection(image,result);
cv::imwrite("cylind_boldt.jpg", result);
return 0;

}

较为关键的是Makefile，因为当前的opencv程序是c++版本，所以需要Makefile稍作修改：
CURDIR=$(shell cd ./; pwd)
TARGET = opencv-test
CROSS_COMPILE = aarch64-himix100-linux-

PPATH = $(CURDIR)
INCLUDE = -I $(CURDIR)/include/ -I $(CURDIR)

LIB += $(PPATH)/lib

LIBS += -L$(LIB) -ldl -lpthread -lm -pthread -lopencv_imgproc -lopencv_core -lopencv_imgcodecs
CFLG = -g -O0 -Wall -lrt -lstdc++ -Wstrict-aliasing

src :=$(wildcard $(shell ls *.cpp))
dir := $(notdir $(src))

CC=$(CROSS_COMPILE)gcc
objs := $(patsubst %cpp, %o, $(src))
$(TARGET):$(objs)
$(CC) $(INCLUDE) $(CFLG) -o $@ $^ $(LIBS)
$(MAKE) --no-print-directory post-build

%.o:%.cpp
$(CC) $(INCLUDE) $(CFLG) -c -o $@ $< $(LIBS)

post-build:
cp $(TARGET) /home/nfs

clean:
rm -f $(TARGET) *.all .o $(PROJECT_SRC_DIR)/.o

将交叉编译好的库文件放到lib目录下，头文件放到include目录下，进行编译，编译出来的程序可以在海思板卡上直接运行。