iOS中基于OpenCV的人脸识别

title: iOS中基于OpenCV的人脸识别
type: categories
date: 2016-11-28 09:45:27
categories: OC

tags: [OpenCV, 人脸识别]

OpenCV 是一个开源的计算机视觉和机器学习库。

与iOS系统自带的识别库一样,对于侧脸的识别很无力。简单记录一下。

相关参考

https://objccn.io/issue-21-9/

http://www.jianshu.com/p/7a055b7eb14c

本文demo

https://github.com/zhangzhaopds/OpenCV_Demo.git

效果图

部署与使用

CocoaPods

pod 'OpenCV', '~> 3.0.0'
使用
#import "OpenCVUtil.h"

#import <opencv2/opencv.hpp>
#import <opencv2/imgproc/types_c.h>
#import <opencv2/imgcodecs/ios.h>

@implementation OpenCVUtil

+ (UIImage *)convertImage: (UIImage *)image {

    // 初始化一个图片的二维矩阵cvImage
    cv::Mat cvImage;

    // 将图片UIImage对象转为Mat对象
    UIImageToMat(image, cvImage);

    if (!cvImage.empty()) {
        cv::Mat gray;

        // 进一步将图片转为灰度显示
        cv::cvtColor(cvImage, gray, CV_RGB2GRAY);

        // 利用搞死滤镜去除边缘
        cv::GaussianBlur(gray, gray, cv::Size(5, 5), 1.2, 1.2);

        // 计算画布
        cv::Mat edges;
        cv::Canny(gray, edges, 0, 50);

        // 使用白色填充
        cvImage.setTo(cv::Scalar::all(225));

        // 修改边缘颜色
        cvImage.setTo(cv::Scalar(0,128,255,255),edges);

        // 将Mat转换为UIImage
        return MatToUIImage(cvImage);
    }

    return nil;
}

+ (NSArray*)facePointDetectForImage:(UIImage*)image{

    static cv::CascadeClassifier faceDetector;

    static dispatch_once_t onceToken;
    dispatch_once(&onceToken, ^{
        // 添加xml文件
        NSString* cascadePath = [[NSBundle mainBundle]
                                 pathForResource:@"haarcascade_frontalface_alt2"
                                 ofType:@"xml"];
        faceDetector.load([cascadePath UTF8String]);
    });

    cv::Mat faceImage;
    UIImageToMat(image, faceImage);

    // 转为灰度
    cv::Mat gray;
    cvtColor(faceImage, gray, CV_BGR2GRAY);

    // 检测人脸并储存
    std::vector<cv::Rect>faces;
    faceDetector.detectMultiScale(gray, faces,1.1,2,CV_HAAR_FIND_BIGGEST_OBJECT,cv::Size(30,30));

    NSMutableArray *array = [NSMutableArray array];

    for(unsigned int i= 0;i < faces.size();i++)
    {
        const cv::Rect& face = faces[i];
        float height = (float)faceImage.rows;
        float width = (float)faceImage.cols;
        CGRect rect = CGRectMake(face.x/width, face.y/height, face.width/width, face.height/height);
        [array addObject:[NSNumber valueWithCGRect:rect]];
    }

    return [array copy];
}

+ (UIImage*)faceDetectForImage:(UIImage*)image {

    static cv::CascadeClassifier faceDetector;

    static dispatch_once_t onceToken;
    dispatch_once(&onceToken, ^{

        NSString* cascadePath = [[NSBundle mainBundle]
                        pathForResource:@"haarcascade_frontalface_alt"
                                 ofType:@"xml"];
        faceDetector.load([cascadePath UTF8String]);
    });

    cv::Mat faceImage;
    UIImageToMat(image, faceImage);

    // 转为灰度
    cv::Mat gray;
    cvtColor(faceImage, gray, CV_BGR2GRAY);

    // 检测人脸并储存
    std::vector<cv::Rect>faces;
    faceDetector.detectMultiScale(gray, faces,1.1,2,0,cv::Size(30,30));

    // 在每个人脸上画一个红色四方形
    for(unsigned int i= 0;i < faces.size();i++)
    {
        const cv::Rect& face = faces[i];
        cv::Point tl(face.x,face.y);
        cv::Point br = tl + cv::Point(face.width,face.height);
        // 四方形的画法
        cv::Scalar magenta = cv::Scalar(255, 0, 0, 255);
        cv::rectangle(faceImage, tl, br, magenta, 11, 8, 0);
    }

    return MatToUIImage(faceImage);
}

+ (UIImage*)circleDetectForImage:(UIImage*)image{
    cv::Mat circleImage,src_gray;
    UIImageToMat(image, circleImage);
    /// Convert it to gray
    cvtColor( circleImage, src_gray, CV_BGR2GRAY );

    /// Reduce the noise so we avoid false circle detection
    GaussianBlur( src_gray, src_gray, cv::Size(9, 9), 2, 2 );

    std::vector<cv::Vec3f> circles;

    /// Apply the Hough Transform to find the circles
    HoughCircles( src_gray, circles, CV_HOUGH_GRADIENT, 1, src_gray.rows/8, 200, 100, 0, 0 );

    /// Draw the circles detected
    for( size_t i = 0; i < circles.size(); i++ )
    {
        cv::Point center(cvRound(circles[i][0]), cvRound(circles[i][1]));
        int radius = cvRound(circles[i][2]);
        // circle center
        circle( circleImage, center, 3, cv::Scalar(0,255,0,255), -1, 8, 0 );
        // circle outline
        circle( circleImage, center, radius, cv::Scalar(0,0,255,255), 3, 8, 0 );
    }

    /// Show your results
    return MatToUIImage(circleImage);

}
@end

调用

- (void)viewDidLoad {
    [super viewDidLoad];
    // Do any additional setup after loading the view, typically from a nib.

    UIImage *img = [UIImage imageNamed:@"888.JPG"];
    _myImgView.image = img;
    _detectImgView.image = [OpenCVUtil faceDetectForImage:img];
}
  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值