How to capture video frames from the camera as images using AV Foundation on iOS

ViewController.swift

//
//  ViewController.swift
//
// Technical Q&A QA1702
// How to capture video frames from the camera as images using AV Foundation on iOS
//

import UIKit
import AVFoundation
import CoreMedia

class ViewController: UIViewController, AVCaptureVideoDataOutputSampleBufferDelegate
{
    let mainGroup = UIStackView()
    let imageView = UIImageView(frame: CGRectZero)

    override func viewDidLoad()
    {
        super.viewDidLoad()

        // 1. Layout Views
        view.addSubview(mainGroup)
        mainGroup.axis = UILayoutConstraintAxis.Vertical
        mainGroup.distribution = UIStackViewDistribution.Fill
        
        mainGroup.addArrangedSubview(imageView)

        imageView.contentMode = UIViewContentMode.ScaleAspectFit
        

        // 2. Create session & configure
        let captureSession = AVCaptureSession()
        captureSession.sessionPreset = AVCaptureSessionPresetMedium


        // 3. set input
        let backCamera = AVCaptureDevice.defaultDeviceWithMediaType(AVMediaTypeVideo)
        do {
            let input = try AVCaptureDeviceInput(device: backCamera)
            
            captureSession.addInput(input)
        }
        catch {
            print("can't access camera")
            return
        }
        
        // although we don't use this, it's required to get captureOutput invoked
        let previewLayer = AVCaptureVideoPreviewLayer(session: captureSession)
        view.layer.addSublayer(previewLayer)


        // 4. set output & configure
        let videoOutput = AVCaptureVideoDataOutput()
        if captureSession.canAddOutput(videoOutput) {
            captureSession.addOutput(videoOutput)
        }
        let queue = dispatch_queue_create("sample buffer delegate", nil)
        videoOutput.setSampleBufferDelegate(self, queue: queue)

        // 5. Specify the pixel format
        videoOutput.videoSettings = [kCVPixelBufferPixelFormatTypeKey : UInt(kCVPixelFormatType_32BGRA)]


        // 6. start running session
        captureSession.startRunning()
    }

    override func viewDidLayoutSubviews()
    {
        let topMargin = topLayoutGuide.length

        mainGroup.frame = CGRect(x: 0, y: topMargin, width: view.frame.width, height: view.frame.height - topMargin).insetBy(dx: 5, dy: 5)
    }

    func captureOutput(captureOutput: AVCaptureOutput!, didOutputSampleBuffer sampleBuffer: CMSampleBuffer!, fromConnection connection: AVCaptureConnection!)
    {

        let image = imageFromSampleBuffer(sampleBuffer)

        dispatch_async(dispatch_get_main_queue())
        {
                print(image.CGImage)
                self.imageView.image = image
        }
    }

    func imageFromSampleBuffer(sampleBuffer: CMSampleBuffer) -> UIImage {
        let imageBuffer:CVImageBuffer! = CMSampleBufferGetImageBuffer(sampleBuffer)

        CVPixelBufferLockBaseAddress(imageBuffer, 0)

        let baseAddress: UnsafeMutablePointer<Void> = CVPixelBufferGetBaseAddress(imageBuffer)
        let bytesPerRow = CVPixelBufferGetBytesPerRow(imageBuffer)
        let width = CVPixelBufferGetWidth(imageBuffer)
        let height = CVPixelBufferGetHeight(imageBuffer)

        let colorSpace = CGColorSpaceCreateDeviceRGB();

        // Create a bitmap graphics context with the sample buffer data
        let context = CGBitmapContextCreate(baseAddress, width, height, 8, bytesPerRow, colorSpace, CGBitmapInfo.ByteOrder32Little.rawValue | CGImageAlphaInfo.PremultipliedFirst.rawValue);
        let quartzImage = CGBitmapContextCreateImage(context)

        CVPixelBufferUnlockBaseAddress(imageBuffer, 0)

        // Create an image object from the Quartz image
        let image = UIImage(CGImage:quartzImage!);

        return (image);
    }
    
    func colorForPixel(image: UIImage, x: Int, y: Int) -> UIColor {
        let cgImage =  image.CGImage

        let imageWidth = CGImageGetWidth(cgImage)
        let imageHeight = CGImageGetHeight(cgImage)
        let bytesPerPixel = CGImageGetBitsPerPixel(cgImage) / 8

        //print("\(imageWidth) x \(imageHeight) x \(bytesPerPixel)")

        let pixelData:CFData! = CGDataProviderCopyData(CGImageGetDataProvider(image.CGImage));
        let data:UnsafePointer<UInt8> = CFDataGetBytePtr(pixelData)

        // CGImage is flipped. So (imageHeight - y) is needed.
        let pixelValue = data + (imageWidth * (imageHeight - y) + x) * bytesPerPixel

        // kCVPixelFormatType_32BGRA
        let blue  = CGFloat(pixelValue[0])
        let green = CGFloat(pixelValue[1])
        let red   = CGFloat(pixelValue[2])
        let alpha = CGFloat(pixelValue[3])

        return UIColor(red: red / 255, green: green / 255, blue: blue / 255, alpha: alpha / 255)
    }
}