opencv常用转换 opencv图像处理

关注

原创最新推荐文章于 2026-04-19 23:29:29 发布 7.4k 阅读 AI 写同款· GEO 优化 ›

python读RAW图片以及大小端转换 float32位大端存储的RAW图片用python读取 infile = "raw_path.raw" img = np.fromfile(infile, dtype='>f') 注意：dtype='float32’默认读取小端存储 float32位大端存储的RAW图片用python转换为小端存储 infile = "raw_path.raw" outfile = "little_endian_raw_path.raw" np.memmap(infile, dtype=np.float32).byt 阅读详情

分配与释放图像空间

分配图像空间:

IplImage* cvCreateImage(CvSize size, int depth, int channels);
 
   size:   cvSize(width,height);
 
   depth: IPL_DEPTH_8U, IPL_DEPTH_8S, IPL_DEPTH_16U,
          IPL_DEPTH_16S, IPL_DEPTH_32S, IPL_DEPTH_32F, IPL_DEPTH_64F
 
   channels: 1, 2, 3 or 4. 
     注意数据为交叉存取.彩色图像的数据编排为b0 g0 r0 b1 g1 r1 ...

举例:

// 分配一个单通道字节图像
IplImage* img1=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1); 
 
// 分配一个三通道浮点图像
IplImage* img2=cvCreateImage(cvSize(640,480),IPL_DEPTH_32F,3);

释放图像空间:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1); 
cvReleaseImage(&img);

复制图像:

IplImage* img1=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1); 
IplImage* img2;
img2=cvCloneImage(img1);

设定/获取兴趣区域:

void   cvSetImageROI(IplImage* image, CvRect rect);
void   cvResetImageROI(IplImage* image);
vRect cvGetImageROI(const IplImage* image);
 
大部分OpenCV函数都支持ROI.

设定/获取兴趣通道:

void cvSetImageCOI(IplImage* image, int coi); // 0=all
int cvGetImageCOI(const IplImage* image);
 
大部分OpenCV函数暂不支持COI.

读取存储图像

从文件中载入图像:

   IplImage* img=0; 
   img=cvLoadImage(fileName);
   if(!img) printf("Could not load image file: %s/n",fileName);
 
   Supported image formats: BMP, DIB, JPEG, JPG, JPE, PNG, PBM, PGM, PPM,
                            SR, RAS, TIFF, TIF

载入图像默认转为3通道彩色图像. 如果不是，则需加flag:

   img=cvLoadImage(fileName,flag);
 
   flag: >0 载入图像转为三通道彩色图像
         =0 载入图像转为单通道灰度图像
         <0 不转换载入图像(通道数与图像文件相同).

图像存储为图像文件:

   if(!cvSaveImage(outFileName,img)) printf("Could not save: %s/n",outFileName);

输入文件格式由文件扩展名决定.

存取图像元素

假设需要读取在i行j列像点的第k通道. 其中, 行数i的范围为[0, height-1], 列数j的范围为[0, width-1], 通道k的范围为[0, nchannels-1].

间接存取: (比较通用, 但效率低, 可读取任一类型图像数据)

对单通道字节图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1);
CvScalar s;
s=cvGet2D(img,i,j); // get the (i,j) pixel value
printf("intensity=%f/n",s.val[0]);
s.val[0]=111;
cvSet2D(img,i,j,s); // set the (i,j) pixel value

对多通道浮点或字节图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_32F,3);
CvScalar s;
s=cvGet2D(img,i,j); // get the (i,j) pixel value
printf("B=%f, G=%f, R=%f/n",s.val[0],s.val[1],s.val[2]);
s.val[0]=111;
s.val[1]=111;
s.val[2]=111;
cvSet2D(img,i,j,s); // set the (i,j) pixel value

直接存取: (效率高, 但容易出错)

对单通道字节图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1);
((uchar *)(img->imageData + i*img->widthStep))[j]=111;

对多通道字节图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,3);
((uchar *)(img->imageData + i*img->widthStep))[j*img->nChannels + 0]=111; // B
((uchar *)(img->imageData + i*img->widthStep))[j*img->nChannels + 1]=112; // G
((uchar *)(img->imageData + i*img->widthStep))[j*img->nChannels + 2]=113; // R

对多通道浮点图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_32F,3);
((float *)(img->imageData + i*img->widthStep))[j*img->nChannels + 0]=111; // B
((float *)(img->imageData + i*img->widthStep))[j*img->nChannels + 1]=112; // G
((float *)(img->imageData + i*img->widthStep))[j*img->nChannels + 2]=113; // R

用指针直接存取 : (在某些情况下简单高效)

对单通道字节图像:

IplImage* img   = cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1);
int height      = img->height;
int width       = img->width;
int step        = img->widthStep/sizeof(uchar);
uchar* data     = (uchar *)img->imageData;
data[i*step+j] = 111;

对多通道字节图像:

IplImage* img   = cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,3);
int height      = img->height;
int width       = img->width;
int step        = img->widthStep/sizeof(uchar);
int channels    = img->nChannels;
uchar* data     = (uchar *)img->imageData;
data[i*step+j*channels+k] = 111;

对单通道浮点图像(假设用4字节调整):

IplImage* img   = cvCreateImage(cvSize(640,480),IPL_DEPTH_32F,3);
int height      = img->height;
int width       = img->width;
int step        = img->widthStep/sizeof(float);
int channels    = img->nChannels;
float * data     = (float *)img->imageData;
data[i*step+j*channels+k] = 111;

使用 c++ wrapper 进行直接存取: (简单高效)

对单/多通道字节图像，多通道浮点图像定义一个 c++ wrapper:

template<class T> class Image
{
   private:
   IplImage* imgp;
   public:
   Image(IplImage* img=0) {imgp=img;}
   ~Image(){imgp=0;}
   void operator=(IplImage* img) {imgp=img;}
   inline T* operator[](const int rowIndx) {
     return ((T *)(imgp->imageData + rowIndx*imgp->widthStep));}
};
 
typedef struct{
   unsigned char b,g,r;
} RgbPixel;
 
typedef struct{
   float b,g,r;
} RgbPixelFloat;
 
typedef Image<RgbPixel>        RgbImage;
typedef Image<RgbPixelFloat>   RgbImageFloat;
typedef Image<unsigned char>   BwImage;
typedef Image<float>           BwImageFloat;

单通道字节图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,1);
BwImage imgA(img);
imgA[i][j] = 111;

多通道字节图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_8U,3);
RgbImage   imgA(img);
imgA[i][j].b = 111;
imgA[i][j].g = 111;
imgA[i][j].r = 111;

多通道浮点图像:

IplImage* img=cvCreateImage(cvSize(640,480),IPL_DEPTH_32F,3);
RgbImageFloat imgA(img);
imgA[i][j].b = 111;
imgA[i][j].g = 111;
imgA[i][j].r = 111;

图像转换

转为灰度或彩色字节图像:

cvConvertImage(src, dst, flags=0);
 
   src = float/byte grayscale/color image
   dst = byte grayscale/color image
   flags = CV_CVTIMG_FLIP      (flip vertically)
           CV_CVTIMG_SWAP_RB   (swap the R and B channels)

转换彩色图像为灰度图像:

使用OpenCV转换函数:

cvCvtColor(cimg,gimg,CV_BGR2GRAY); // cimg -> gimg

直接转换:

for(i=0;i<cimg->height;i++) for(j=0;j<cimg->width;j++) 
   gimgA[i][j]= (uchar)(cimgA[i][j].b*0.114 + 
                        cimgA[i][j].g*0.587 + 
                        cimgA[i][j].r*0.299);

颜色空间转换:

cvCvtColor(src,dst,code); // src -> dst
 
   code     = CV_<X>2<Y>
   <X>/<Y> = RGB, BGR, GRAY, HSV, YCrCb, XYZ, Lab, Luv, HLS
 
e.g.: CV_BGR2GRAY, CV_BGR2HSV, CV_BGR2Lab

绘图命令

画长方体:

// 用宽度为1的红线在(100,100)与(200,200)之间画一长方体
cvRectangle(img, cvPoint(100,100), cvPoint(200,200), cvScalar(255,0,0), 1);

画圆:

// 在(100,100)处画一半径为20的圆,使用宽度为1的绿线
cvCircle(img, cvPoint(100,100), 20, cvScalar(0,255,0), 1);

画线段:

// 在(100,100)与(200,200)之间画绿色线段，宽度为1
cvLine(img, cvPoint(100,100), cvPoint(200,200), cvScalar(0,255,0), 1);

画一组线段:

CvPoint   curve1[]={10,10,   10,100,   100,100,   100,10};
CvPoint   curve2[]={30,30,   30,130,   130,130,   130,30,   150,10};
CvPoint* curveArr[2]={curve1, curve2};
int       nCurvePts[2]={4,5};
int       nCurves=2;
int       isCurveClosed=1;
int       lineWidth=1;
 
cvPolyLine(img,curveArr,nCurvePts,nCurves,isCurveClosed,cvScalar(0,255,255),lineWidth);

画内填充色的多边形:

cvFillPoly(img,curveArr,nCurvePts,nCurves,cvScalar(0,255,255));

添加文本:

CvFont font;
double hScale=1.0;
double vScale=1.0;
int     lineWidth=1;
cvInitFont(&font,CV_FONT_HERSHEY_SIMPLEX|CV_FONT_ITALIC, hScale,vScale,0,lineWidth);
 
cvPutText (img,"My comment",cvPoint(200,400), &font, cvScalar(255,255,0));

Other possible fonts:

CV_FONT_HERSHEY_SIMPLEX, CV_FONT_HERSHEY_PLAIN,
CV_FONT_HERSHEY_DUPLEX, CV_FONT_HERSHEY_COMPLEX,
CV_FONT_HERSHEY_TRIPLEX, CV_FONT_HERSHEY_COMPLEX_SMALL,
CV_FONT_HERSHEY_SCRIPT_SIMPLEX, CV_FONT_HERSHEY_SCRIPT_COMPLEX,

收录于