OpenCV是什么

OpenCV是一个基于BSD许可（开源）发行的跨平台计算机视觉库，可以运行在Linux、Windows、Android和Mac OS操作系统上。它轻量级而且高效——由一系列 C 函数和少量 C++ 类构成，同时提供了Python、Ruby、MATLAB等语言的接口，实现了图像处理和计算机视觉方面的很多通用算法。

如何在iOS开发中使用OpenCV

1.在OpenCV的官网下载iOS端的framework 地址：http://opencv.org/releases.html
2.将下载好的framework拖进Xcode，因为OpenCV是C++编写，所以需要在代码里面进行一些修改。将引入OpenCV头文件的.m文件后缀改为.mm。

人脸匹配是怎样实现的？

将包含人脸的图片作为基准图，然后与匹配图片做直方图比对

实现代码

#ifdef __cplusplus
#import <opencv2/opencv.hpp>
#endif
#import <opencv2/videoio/cap_ios.h>
#import <opencv2/imgproc/imgproc_c.h>
#import <opencv2/core/core_c.h>
#import <opencv2/features2d/features2d.hpp>


@interface ViewController ()
@property(nonatomic,strong) UIImageView* imageView;
@property(nonatomic,strong) UIImageView* imageView1;
@end

@implementation ViewController



- (void)viewDidLoad {
    [super viewDidLoad];
    //识别的图片
    UIImage *mImage =  [UIImage imageNamed:@"31.jpeg"];
    IplImage *srcIpl = [self convertToIplImage:mImage];
    IplImage *dscIpl = cvCreateImage(cvGetSize(srcIpl), srcIpl->depth, 1);
    IplImage *dscIplNew = cvCreateImage(cvGetSize(srcIpl),  IPL_DEPTH_8U, 3);
    cvCvtColor(dscIpl, dscIplNew, CV_GRAY2BGR);
    
    //基准图
    UIImage *mImage1 =  [UIImage imageNamed:@"32.jpeg"];
    IplImage *srcIpl1 = [self convertToIplImage:mImage1];
    IplImage *dscIpl1 = cvCreateImage(cvGetSize(srcIpl1), srcIpl1 ->depth, 1);
    IplImage *dscIplNew1 = cvCreateImage(cvGetSize(srcIpl1), IPL_DEPTH_8U, 3);
    cvCvtColor(dscIpl1, dscIplNew1, CV_GRAY2BGR);
    //基准图2
    UIImage *tempImage = [UIImage imageNamed:@"30.jpeg"];
    IplImage *iplTempImage = [self convertToIplImage:tempImage];
    BOOL tf=[self ComparePPKImage:srcIpl withAnotherImage:srcIpl1 withTempleImage:iplTempImage];
    if (tf) {
        printf("匹配成功\n");
    }else
    {
        printf("匹配失败\n");
    }
}

//图片匹配
-(BOOL)ComparePPKImage:(IplImage*)mIplImage withAnotherImage:(IplImage*)mIplImage1 withTempleImage:(IplImage*)mTempleImage
{
    //第一次模板标记
    CvPoint minLoc =[self CompareTempleImage:mTempleImage withImage:mIplImage];
    if (minLoc.x==mIplImage->width || minLoc.y==mIplImage->height) {
        printf("第一个图片的模板标记失败\n");
        return false;
    }
    //第二次模板标记
    CvPoint minLoc1 =[self CompareTempleImage:mTempleImage withImage:mIplImage1];
    if (minLoc1.x==mIplImage1->width || minLoc1.y==mIplImage1->height) {
        printf("第二个图片的模板标记失败\n");
        return false;
    }
    //裁切图片
    IplImage *cropImage,*cropImage1;
    cropImage =[self cropIplImage:mIplImage withStartPoint:minLoc withWidth:mTempleImage->width withHeight:mTempleImage->height];
    cropImage1=[self cropIplImage:mIplImage1 withStartPoint:minLoc1 withWidth:mTempleImage->width withHeight:mTempleImage->height];
    self.imageView.image=[self convertToUIImage:cropImage];
    self.imageView1.image=[self convertToUIImage:cropImage1];
    double rst = [self CompareHist:cropImage withParam2:cropImage1];
    if (rst<0.18) {
        return true;
    }
    else
    {
        return false;
    }
}

/// 基于模板图片的标记识别
-(CvPoint)CompareTempleImage:(IplImage*)templeIpl withImage:(IplImage*)mIplImage
{
    IplImage *src = mIplImage;
    IplImage *templat = templeIpl;
    IplImage *result;
    int srcW, srcH, templatW, templatH, resultH, resultW;
    srcW = src->width;
    srcH = src->height;
    templatW = templat->width;
    templatH = templat->height;
    resultW = srcW - templatW + 1;
    resultH = srcH - templatH + 1;
    result = cvCreateImage(cvSize(resultW, resultH), 32, 1);
    cvMatchTemplate(src, templat, result, CV_TM_SQDIFF);
    double minValue, maxValue;
    CvPoint minLoc, maxLoc;
    cvMinMaxLoc(result, &minValue, &maxValue, &minLoc, &maxLoc);
    if (minLoc.y+templatH>srcH || minLoc.x+templatW>srcW) {
        printf("未找到标记图片\n");
        minLoc.x=srcW;
        minLoc.y=srcH;
    }
    return minLoc;
}

    // Do any additional setup after loading the view, typically from a nib.
/// UIImage类型转换为IPlImage类型
-(IplImage*)convertToIplImage:(UIImage*)image
{
    CGImageRef imageRef = image.CGImage;
    CGColorSpaceRef colorSpace = CGColorSpaceCreateDeviceRGB();
    IplImage *iplImage = cvCreateImage(cvSize(image.size.width, image.size.height), IPL_DEPTH_8U, 4);
    CGContextRef contextRef = CGBitmapContextCreate(iplImage->imageData, iplImage->width, iplImage->height, iplImage->depth, iplImage->widthStep, colorSpace, kCGImageAlphaPremultipliedLast|kCGBitmapByteOrderDefault);
    CGContextDrawImage(contextRef, CGRectMake(0, 0, image.size.width, image.size.height), imageRef);
    CGContextRelease(contextRef);
    CGColorSpaceRelease(colorSpace);
    IplImage *ret = cvCreateImage(cvGetSize(iplImage), IPL_DEPTH_8U, 3);
    cvCvtColor(iplImage, ret, CV_RGB2BGR);
    cvReleaseImage(&iplImage);
    return ret;
}

/// IplImage类型转换为UIImage类型
-(UIImage*)convertToUIImage:(IplImage*)image
{
    cvCvtColor(image, image, CV_BGR2RGB);
    CGColorSpaceRef colorSpace = CGColorSpaceCreateDeviceRGB();
    NSData *data = [NSData dataWithBytes:image->imageData length:image->imageSize];
    CGDataProviderRef provider = CGDataProviderCreateWithCFData((CFDataRef)data);
    CGImageRef imageRef = CGImageCreate(image->width, image->height, image->depth, image->depth * image->nChannels, image->widthStep, colorSpace, kCGImageAlphaNone | kCGBitmapByteOrderDefault, provider, NULL, false, kCGRenderingIntentDefault);
    UIImage *ret = [UIImage imageWithCGImage:imageRef];
    CGImageRelease(imageRef);
    CGDataProviderRelease(provider);
    CGColorSpaceRelease(colorSpace);
    return ret;
}
-(IplImage*)cropIplImage:(IplImage*)srcIpl withStartPoint:(CvPoint)mPoint withWidth:(int)width withHeight:(int)height
{
    //裁剪后的图片
    IplImage *cropImage;
    cvSetImageROI(srcIpl, cvRect(mPoint.x, mPoint.y, width, height));
    cropImage = cvCreateImage(cvGetSize(srcIpl), IPL_DEPTH_8U, 3);
    cvCopy(srcIpl, cropImage);
    cvResetImageROI(srcIpl);
    return cropImage;
}

// 多通道彩色图片的直方图比对
-(double)CompareHist:(IplImage*)image1 withParam2:(IplImage*)image2
{
    int hist_size = 256;
    IplImage *gray_plane = cvCreateImage(cvGetSize(image1), 8, 1);
    cvCvtColor(image1, gray_plane, CV_BGR2GRAY);
    float range[] = {0,255};  //灰度级的范围
    float* ranges[]={range};
    CvHistogram *gray_hist = cvCreateHist(1, &hist_size, CV_HIST_ARRAY,ranges,1);
    cvCalcHist(&gray_plane, gray_hist);
    
    IplImage *gray_plane2 = cvCreateImage(cvGetSize(image2), 8, 1);
    cvCvtColor(image2, gray_plane2, CV_BGR2GRAY);
    CvHistogram *gray_hist2 = cvCreateHist(1, &hist_size, CV_HIST_ARRAY,ranges,1);
    cvCalcHist(&gray_plane2, gray_hist2);
    double rst =cvCompareHist(gray_hist, gray_hist2, CV_COMP_BHATTACHARYYA);
    printf("对比结果=%f\n",rst);
    return rst;
}

注意事项

运行时出现下图错误，你会发现报错的是
enum { NO, GAIN, GAIN_BLOCKS };
这句代码，不要着急，用这句代码替换即可
enum { NO_EXPOSURE_COMPENSATOR = 0, GAIN, GAIN_BLOCKS };

错误截图

缺陷

该种方法对图片相似度要求较高，当背景差异较大或者干扰因素较多时，无法匹配成功

demo地址

https://github.com/Fairy-happy/Picture-matchingWithOpenCV

参考链接

http://blog.csdn.net/wanggsx918/article/details/23833425