将YUV_420_888转换为JPEG并保存文件会导致图像失真 [英] Converting YUV_420_888 to JPEG and saving file results distorted image

问题描述

我使用了

如果我已将 ImageReader 更改为使用 ImageFormat.JPEG 而是通过执行以下更改在 Camera2 ：

  mPreviewImageReader = ImageReader.newInstance（previewSize.getWidth（ ），
 previewSize.getHeight（），ImageFormat.JPEG，/ * maxImages * / 2）; 
 mCamera.createCaptureSession（Arrays.asList（surface，mPreviewImageReader.getSurface（）），
 mSessionCallback，null）; 
  

图像正常运行而没有任何失真但帧速率显着下降且视图开始滞后。因此我认为 ImageUtil 类没有正确转换。

解决方案

解决方案由@ volodymyr-kulyk提供的没有考虑图像中平面的行间距。下面的代码是诀窍（图像是 android.media.Image 类型）：

  data = NV21toJPEG（YUV420toNV21（image），image.getWidth（），image.getHeight（），100）; 
  

实施：

  private static byte [] NV21toJPEG（byte [] nv21，int width，int height，int quality）{
 ByteArrayOutputStream out = new ByteArrayOutputStream（）; 
 YuvImage yuv = new YuvImage（nv21，ImageFormat.NV21，width，height，null）; 
 yuv.compressToJpeg（new Rect（0,0，width，height），quality，out）; 
返回out.toByteArray（）; 
} 
 
 private static byte [] YUV420toNV21（图片图片）{
 Rect crop = image.getCropRect（）; 
 int format = image.getFormat（）; 
 int width = crop.width（）; 
 int height = crop.height（）; 
 Image.Plane [] planes = image.getPlanes（）; 
 byte [] data = new byte [width * height * ImageFormat.getBitsPerPixel（format）/ 8]; 
 byte [] rowData = new byte [planes [0] .getRowStride（）]; 
 
 int channelOffset = 0; 
 int outputStride = 1; 
 for（int i = 0; i< planes.length; i ++）{
 switch（i）{
 case 0：
 channelOffset = 0; 
 outputStride = 1; 
休息; 
案例1：
 channelOffset = width * height + 1; 
 outputStride = 2; 
休息; 
案例2：
 channelOffset = width * height; 
 outputStride = 2; 
休息; 
} 
 
 ByteBuffer buffer = planes [i] .getBuffer（）; 
 int rowStride = planes [i] .getRowStride（）; 
 int pixelStride = planes [i] .getPixelStride（）; 
 
 int shift =（i == 0）？ 0：1; 
 int w = width>>转移; 
 int h = height>>转移; 
 buffer.position（rowStride *（crop.top>> shift）+ pixelStride *（crop.left>> shift））; 
 for（int row = 0; row< h; row ++）{
 int length; 
 if（pixelStride == 1&& outputStride == 1）{
 length = w; 
 buffer.get（data，channelOffset，length）; 
 channelOffset + = length; 
} else {
 length =（w  -  1）* pixelStride + 1; 
 buffer.get（rowData，0，length）; 
 for（int col = 0; col< w; col ++）{
 data [channelOffset] = rowData [col * pixelStride]; 
 channelOffset + = outputStride; 
} 
} 
 if（row< h  -  1）{
 buffer.position（buffer.position（）+ rowStride  -  length）; 
} 
} 
} 
返回数据; 
} 
  

方法来自以下链接。

I've used the ImageUtil class provided in https://stackoverflow.com/a/40152147/2949966 within my git repo: https://github.com/ahasbini/cameraview/tree/camera_preview_imp (note the implementation is in camera_preview_imp branch) to implement a frame preview callback. An ImageReader is set to preview frames in the ImageFormat.YUV_420_888 format which will be converted into ImageFormat.JPEG using the ImageUtil class and send it to the frame callback. The demo app saves a frame from the callback to a file every 50 frames. All of the saved frame images are coming out distorted similar to below:

If I've changed the ImageReader to use ImageFormat.JPEG instead by doing the following changes in Camera2:

mPreviewImageReader = ImageReader.newInstance(previewSize.getWidth(),
    previewSize.getHeight(), ImageFormat.JPEG, /* maxImages */ 2);
mCamera.createCaptureSession(Arrays.asList(surface, mPreviewImageReader.getSurface()),
    mSessionCallback, null);

the image is coming properly without any distortions however the frame rate drops significantly and the view starts to lag. Hence I believe the ImageUtil class is not converting properly.

解决方案

Solution provided by @volodymyr-kulyk does not take into consideration the row stride of the planes within the image. Below code does the trick (image is of android.media.Image type):

data = NV21toJPEG(YUV420toNV21(image), image.getWidth(), image.getHeight(), 100);

And the implementations:

private static byte[] NV21toJPEG(byte[] nv21, int width, int height, int quality) {
    ByteArrayOutputStream out = new ByteArrayOutputStream();
    YuvImage yuv = new YuvImage(nv21, ImageFormat.NV21, width, height, null);
    yuv.compressToJpeg(new Rect(0, 0, width, height), quality, out);
    return out.toByteArray();
}

private static byte[] YUV420toNV21(Image image) {
    Rect crop = image.getCropRect();
    int format = image.getFormat();
    int width = crop.width();
    int height = crop.height();
    Image.Plane[] planes = image.getPlanes();
    byte[] data = new byte[width * height * ImageFormat.getBitsPerPixel(format) / 8];
    byte[] rowData = new byte[planes[0].getRowStride()];

    int channelOffset = 0;
    int outputStride = 1;
    for (int i = 0; i < planes.length; i++) {
        switch (i) {
            case 0:
                channelOffset = 0;
                outputStride = 1;
                break;
            case 1:
                channelOffset = width * height + 1;
                outputStride = 2;
                break;
            case 2:
                channelOffset = width * height;
                outputStride = 2;
                break;
        }

        ByteBuffer buffer = planes[i].getBuffer();
        int rowStride = planes[i].getRowStride();
        int pixelStride = planes[i].getPixelStride();

        int shift = (i == 0) ? 0 : 1;
        int w = width >> shift;
        int h = height >> shift;
        buffer.position(rowStride * (crop.top >> shift) + pixelStride * (crop.left >> shift));
        for (int row = 0; row < h; row++) {
            int length;
            if (pixelStride == 1 && outputStride == 1) {
                length = w;
                buffer.get(data, channelOffset, length);
                channelOffset += length;
            } else {
                length = (w - 1) * pixelStride + 1;
                buffer.get(rowData, 0, length);
                for (int col = 0; col < w; col++) {
                    data[channelOffset] = rowData[col * pixelStride];
                    channelOffset += outputStride;
                }
            }
            if (row < h - 1) {
                buffer.position(buffer.position() + rowStride - length);
            }
        }
    }
    return data;
}

Method was gotten from the following link.

这篇关于将YUV_420_888转换为JPEG并保存文件会导致图像失真的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！