如何在SwiftUI视图中获取AVKit计算的亮度值？

要让SwiftUI视图获取VideoDelegate中计算出的亮度值，核心是让代理类能把数据传递给VideoStream这个ObservableObject，再通过@Published属性让视图自动响应更新。以下是具体实现方案：

修改步骤与代码 #

1. 给VideoDelegate添加回调闭包 #

在代理类中定义一个闭包，用于传递计算好的亮度值，同时优化原代码中的强制解包逻辑避免崩溃：

class VideoDelegate: NSObject, AVCaptureVideoDataOutputSampleBufferDelegate {
    // 定义回调闭包，用于传递亮度值
    var luminosityUpdateHandler: ((Double) -> Void)?
    
    func captureOutput(_ output: AVCaptureOutput, didOutput sampleBuffer: CMSampleBuffer, from connection: AVCaptureConnection) {
        // 读取帧的EXIF元数据
        let rawMetadata = CMCopyDictionaryOfAttachments(allocator: nil, target: sampleBuffer, attachmentMode: CMAttachmentMode(kCMAttachmentMode_ShouldPropagate))
        let metadata = CFDictionaryCreateMutableCopy(nil, 0, rawMetadata) as NSMutableDictionary
        guard let exifData = metadata.value(forKey: "{Exif}") as? NSMutableDictionary else {
            return
        }
        
        // 安全解析EXIF参数，替换原强制解包
        guard let fNumber = exifData["FNumber"] as? Double,
              let exposureTime = exifData["ExposureTime"] as? Double,
              let isoArray = exifData["ISOSpeedRatings"] as? NSArray,
              let isoSpeed = isoArray.firstObject as? Double else {
            return
        }
        
        let calibrationConstant: Double = 50
        let luminosity = (calibrationConstant * fNumber * fNumber) / (exposureTime * isoSpeed)
        
        // 切换到主线程调用回调，保证UI更新安全
        DispatchQueue.main.async {
            self.luminosityUpdateHandler?(luminosity)
        }
    }
}

2. 在VideoStream中绑定回调并持有代理实例 #

修改VideoStream，创建代理时绑定闭包更新@Published属性，同时持有代理实例避免被提前释放：

class VideoStream: ObservableObject {
    
    @Published var luminosityReading: Double = 0.0
    var session: AVCaptureSession!
    // 持有代理实例，防止被自动释放
    private var videoDelegate: VideoDelegate!
    
    init() {
        authorizeCapture()
    }
    
    func authorizeCapture() {
        // 权限申请逻辑，完成后调用beginCapture()
    }
    
    func beginCapture() {        
        session = AVCaptureSession()
        session.beginConfiguration()
        let videoDevice = bestDevice() // 自定义的设备选择逻辑
        
        guard
            let videoDeviceInput = try? AVCaptureDeviceInput(device: videoDevice),
            session.canAddInput(videoDeviceInput)
        else {
            print("Camera selection failed")
            return
        }
        
        let videoOutput = AVCaptureVideoDataOutput()
        guard session.canAddOutput(videoOutput) else {
            print("Error creating video output")
            return
        }
        
        session.sessionPreset = .high
        session.addOutput(videoOutput)
        
        let queue = DispatchQueue(label: "VideoFrameQueue")
        videoDelegate = VideoDelegate()
        // 绑定回调，更新亮度值到Published属性
        videoDelegate.luminosityUpdateHandler = { [weak self] luminosity in
            self?.luminosityReading = luminosity
        }
        videoOutput.setSampleBufferDelegate(videoDelegate, queue: queue)
        
        session.commitConfiguration()
        session.startRunning()
    }
}

3. ContentView保持原有逻辑即可 #

原视图代码不需要修改，@StateObject会自动监听luminosityReading的变化并更新UI：

struct ContentView: View {
    @StateObject var videoStream = VideoStream()
    
    var body: some View {
        Text("\(videoStream.luminosityReading) ?? Detecting...")
            .padding()
    }
}

关键注意事项 #

• 必须在主线程更新@Published属性：captureOutput方法在后台队列执行，直接更新会导致UI异常，所以用DispatchQueue.main.async切换线程。
• 持有代理实例：如果不把VideoDelegate赋值给VideoStream的属性，创建的代理会被立即释放，无法收到回调。
• 安全解析数据：用guard替换强制解包，避免EXIF数据缺失时崩溃。

内容的提问来源于stack exchange，提问作者John Harrington