iOS tutorial 2：用Core Image进行面部识别

参考：Face Detection in iOS Using Core Image

面部识别API不仅可以是识别面部，也可识别面部的特殊细节，例如微笑甚至眨眼睛。

建立初始项目

原文建好了初始项目，我自己新建了初始项目

• 新建项目Detector
• 删除IB中原本View Controller Scene。
• 拖动UITabBarController到IB中，得到三个Scene。选择UITabBarController的Is Initial View Controller，使其作为初始控制器。
• 修改Item 1的title和其Bar Item都为Photo，修改其Class为ViewController。
• 向Assets中添加几张人物图片
• 想Photo Scene中添加一个Image View，Content Mode改为Aspect Fit，选择一个图片。在ViewController添加图片对应@IBOutlet：
  @IBOutlet var personPic: UIImageView!
• 选中Item 2，点击菜单栏EDitor > Embed In > Navigation Controller，新生成一个与之关联的Scene。
• 新建CameraViewController类，继承至UIViewController。修改上面生成的Scene的Class属性为CameraViewController。
• 拖动一个UIBarButtonItem到Camera View Controller Scene的UINavigationItem的右边，并选择System Item为Camera
• 在CameraViewController中建立outlet和Action

识别照片的面部

• 在 ViewController.swift中引入CoreImage:
  import CoreImage
• 在 ViewController.swift中添加函数detect():

func detect() {
    // 1    
    guard let personciImage = CIImage(image: personPic.image!) else {
        return
    }
    // 2 
    let accuracy = [CIDetectorAccuracy: CIDetectorAccuracyHigh]
    let faceDetector = CIDetector(ofType: CIDetectorTypeFace, context: nil, options: accuracy)
    let faces = faceDetector?.features(in: personciImage)
    
    // 3 
    for face in faces as! [CIFaceFeature] {
        
        print("Found bounds are \(face.bounds)")
        
        let faceBox = UIView(frame: face.bounds)
        
        faceBox.layer.borderWidth = 3
        faceBox.layer.borderColor = UIColor.red.cgColor
        faceBox.backgroundColor = UIColor.clear
        personPic.addSubview(faceBox)
        // 4
        if face.hasLeftEyePosition {
            print("Left eye bounds are \(face.leftEyePosition)")
        }
        
        if face.hasRightEyePosition {
            print("Right eye bounds are \(face.rightEyePosition)")
        }
    }
}

• 1 根据UIImage获取CoreImage中图片对象。guard与if功能类似，区别可查看以撸代码的形式学习Swift-5：Control Flow的6 guard 与 if。
• 2 初始化检测器CIDetector， accuray是检查器配置选项，表示精确度；因为CIDetector可以进行几种类型的检测，所以CIDetectorTypeFace用来表示面部检测；features方法返回具体的检测结果
• 3 给每个检测到的脸添加红色框
• 4 检测是否有左眼位置

• 在viewDidLoad中添加 detect()，运行结果类似：

打印结果，显示检测到的面部位置是不对的：
Found bounds are (177.0, 416.0, 380.0, 380.0)
这是因为UIKit的坐标系统与Core Image的坐标系统是不同的：

• 把Core Image的坐标系统转换为UIKit的坐标系统，修改detect()为：

func detect() {
        
    guard let personciImage = CIImage(image: personPic.image!) else {
        return
    }
    
    let accuracy = [CIDetectorAccuracy: CIDetectorAccuracyHigh]
    let faceDetector = CIDetector(ofType: CIDetectorTypeFace, context: nil, options: accuracy)
    let faces = faceDetector?.features(in: personciImage)
    //
    let ciImageSize = personciImage.extent.size
    var transform = CGAffineTransform(scaleX: 1, y: -1)
    transform = transform.translatedBy(x: 0, y: -ciImageSize.height)
    
    for face in faces as! [CIFaceFeature] {
        
        print("Found bounds are \(face.bounds)")
        
        // Apply the transform to convert the coordinates
        var faceViewBounds = face.bounds.applying(transform)
        
        // Calculate the actual position and size of the rectangle in the image view
        let viewSize = personPic.bounds.size
        let scale = min(viewSize.width / ciImageSize.width,
                        viewSize.height / ciImageSize.height)
        let offsetX = (viewSize.width - ciImageSize.width * scale) / 2
        let offsetY = (viewSize.height - ciImageSize.height * scale) / 2
        
        faceViewBounds = faceViewBounds.applying(CGAffineTransform(scaleX: scale, y: scale))
        faceViewBounds.origin.x += offsetX
        faceViewBounds.origin.y += offsetY
        
        let faceBox = UIView(frame: faceViewBounds)
        
        faceBox.layer.borderWidth = 3
        faceBox.layer.borderColor = UIColor.red.cgColor
        faceBox.backgroundColor = UIColor.clear
        personPic.addSubview(faceBox)
        
        if face.hasLeftEyePosition {
            print("Left eye bounds are \(face.leftEyePosition)")
        }
        
        if face.hasRightEyePosition {
            print("Right eye bounds are \(face.rightEyePosition)")
        }
    }
}

运行可看到正确识别位置：

相机拍照后的脸部识别

之前是项目中照片识别，现在是拍完照再识别，原理是相同的，就是多一个拍完照，取照片的过程。

• 更新CameraViewController类的代码

// 1
class CameraViewController: UIViewController, UIImagePickerControllerDelegate, UINavigationControllerDelegate {

    @IBOutlet var imageView: UIImageView!
    // 2
    let imagePicker = UIImagePickerController()
    
    override func viewDidLoad() {
        super.viewDidLoad()
        imagePicker.delegate = self
    }
    
    @IBAction func takePhoto(_ sender: AnyObject) {
        // 3
        if !UIImagePickerController.isSourceTypeAvailable(.camera) {
            return
        }
        imagePicker.allowsEditing = false
        imagePicker.sourceType = .camera
        present(imagePicker, animated: true, completion: nil)
    }
    // 4
    //MARK: -UIImagePickerControllerDelegate
    func imagePickerController(_ picker: UIImagePickerController, didFinishPickingMediaWithInfo info: [String : Any]) {
        if let pickedImage = info[UIImagePickerControllerOriginalImage] as? UIImage {
            imageView.contentMode = .scaleAspectFit
            imageView.image = pickedImage
        }
        
        dismiss(animated: true, completion: nil)
        self.detect()
        
    }
    // 5 
    func imagePickerControllerDidCancel(_ picker: UIImagePickerController) {
        dismiss(animated: true, completion: nil)
    }
}

• 1 实现UIImagePickerControllerDelegate协议，用于拍照相关代理。
• 2 初始化UIImagePickerController。UIImagePickerController是照相或摄影界面和功能管理的类。
• 3 判断设备照相机是否可用。
• 4 实现一个UIImagePickerControllerDelegate中的代理方法，当拍摄完备确实使用照片时调用。
• 5 也是UIImagePickerControllerDelegate中的代理方法，取消拍摄时调用。

• 添加detect()代码，与ViewController中不同的是，不用红色框框处识别出的面部，而是识别出面部的细节，并用UIAlertController弹出显示。

func detect() {
        let imageOptions = NSDictionary(object: NSNumber(value: 5) as NSNumber, forKey: CIDetectorImageOrientation as NSString)
        let personciImage = CIImage(cgImage: imageView.image!.cgImage!)
        let accuracy = [CIDetectorAccuracy: CIDetectorAccuracyHigh]
        let faceDetector = CIDetector(ofType: CIDetectorTypeFace, context: nil, options: accuracy)
        let faces = faceDetector?.features(in: personciImage, options: imageOptions as? [String : AnyObject])
        
        if let face = faces?.first as? CIFaceFeature {
            print("found bounds are \(face.bounds)")
            
            var message = "有个脸"
            
            if face.hasSmile {
                print("脸是笑的")
                message += ",脸是笑的"
            }
            if face.hasMouthPosition {
                print("有嘴唇")
                message += ",有嘴唇"
            }
            
            if face.hasLeftEyePosition {
                print("左眼镜的位置是 \(face.leftEyePosition)")
                message += ",左眼镜的位置是 \(face.leftEyePosition)"
            }
            
            if face.hasRightEyePosition {
                print("右眼镜的位置是 \(face.rightEyePosition)")
                message += ",右眼镜的位置是 \(face.rightEyePosition)"
            }
            
            let alert = UIAlertController(title: "嘿嘿", message: message, preferredStyle: .alert)
            alert.addAction(UIAlertAction(title: "OK", style: .default, handler: nil))
            self.present(alert, animated: true, completion: nil)
            
        } else {
            let alert = UIAlertController(title: "没脸了", message: "没有检测到脸", preferredStyle: .alert)
            alert.addAction(UIAlertAction(title: "OK", style: .default, handler: nil))
            self.present(alert, animated: true, completion: nil)
        }        
}

运行就可以识别照片的面部具体细节
CIFaceFeature还提供了其他很多面部细节：

        open var hasLeftEyePosition: Bool { get }

        open var leftEyePosition: CGPoint { get }

        open var hasRightEyePosition: Bool { get }

        open var rightEyePosition: CGPoint { get }

        open var hasMouthPosition: Bool { get }

        open var mouthPosition: CGPoint { get }

        open var hasTrackingID: Bool { get }

        open var trackingID: Int32 { get }

        open var hasTrackingFrameCount: Bool { get }

        open var trackingFrameCount: Int32 { get }

        open var hasFaceAngle: Bool { get }

        open var faceAngle: Float { get }

        open var hasSmile: Bool { get }

        open var leftEyeClosed: Bool { get }

        open var rightEyeClosed: Bool { get }     

代码

Detector