Android OpenCV使用3_使用OpenCV并进行人脸检测，转竖屏

1.使用OpenCV并进行人脸检测

参考前文，即可完成横屏的人脸检测，地址如下：
Android OpenCV使用2_使用OpenCV并进行人脸检测
https://blog.csdn.net/u013370255/article/details/107252777

2.横屏转竖屏研究

旋转预览获取到的Mat，关键代码MatRotate.matRotateClockWise270(mRgba,mRgba);

		//openCV预览界面监听
        mOpenCvCameraView.setCvCameraViewListener(new CameraBridgeViewBase.CvCameraViewListener2() {
            @Override
            public void onCameraViewStarted(int width, int height) {
            }

            @Override
            public void onCameraViewStopped() {
                mRgba.release();
                mGray.release();
            }

            @Override
            public Mat onCameraFrame(CameraBridgeViewBase.CvCameraViewFrame inputFrame) {
                mRgba = inputFrame.rgba();
                mGray = inputFrame.gray();

				//旋转Mat
                MatRotate.matRotateClockWise270(mRgba,mRgba);

                return mRgba;
            }
        });

新建类MatRotate，封装了Mat旋转方法

public class MatRotate {
    public static void matRotateClockWise90(Mat old, Mat newOne) {
        if (!old.empty()) {
            Core.transpose(old, newOne);
            Core.flip(old, newOne, 1);
        }
    }

    public static void matRotateClockWise180(Mat old, Mat newOne) {
        if (!old.empty()) {
            Core.flip(old, newOne, 0);
            Core.flip(old, newOne, 1);
        }
    }

    public static void matRotateClockWise270(Mat old, Mat newOne) {
        if (!old.empty()) {
            Core.transpose(old, newOne);
            Core.flip(old, newOne, 0);
        }
    }

    public static void myRotateAntiClockWise90(Mat old, Mat newOne) {
        if (!old.empty()) {
            Core.transpose(old, newOne);
            Core.flip(old, newOne, 0);
        }
    }
}

修改OpenCV库源码文件org.opencv.android.CameraBridgeViewBase
找到其中的AllocateCache方法，修改为：

    // NOTE: On Android 4.1.x the function must be called before SurfaceTexture constructor!
    protected void AllocateCache()
    {
        mCacheBitmap = Bitmap.createBitmap( mFrameHeight,mFrameWidth, Bitmap.Config.ARGB_8888);
    }

AllocateCache方法修改需要根据mat旋转情况而定，旋转0或180则不需要修改，旋转90或270则需要调换长宽变量。
通过deliverAndDrawFrame方法的代码可以了解，其从摄像头监听返回值获取到Mat，然后转换为bitmap,再呈现到预览界面，即mCacheBitmap变量，旋转90或270后与摄像头监听返回值Mat长宽不同则无法转换从而会报错或者黑屏退出。

找到其中的deliverAndDrawFrame方法，修改为：

/**
     * This method shall be called by the subclasses when they have valid
     * object and want it to be delivered to external client (via callback) and
     * then displayed on the screen.
     * @param frame - the current frame to be delivered
     */
    protected void deliverAndDrawFrame(CvCameraViewFrame frame) {
        Mat modified;

        if (mListener != null) {
            modified = mListener.onCameraFrame(frame);
        } else {
            modified = frame.rgba();
        }

        boolean bmpValid = true;
        if (modified != null) {
            try {
                Utils.matToBitmap(modified, mCacheBitmap);
            } catch(Exception e) {
                Log.e(TAG, "Mat type: " + modified);
                Log.e(TAG, "Bitmap type: " + mCacheBitmap.getWidth() + "*" + mCacheBitmap.getHeight());
                Log.e(TAG, "Utils.matToBitmap() throws an exception: " + e.getMessage());
                bmpValid = false;
            }
        }

        if (bmpValid && mCacheBitmap != null) {
            Canvas canvas = getHolder().lockCanvas();
            if (canvas != null) {
                canvas.drawColor(0, android.graphics.PorterDuff.Mode.CLEAR);
                if (BuildConfig.DEBUG)
                    Log.d(TAG, "mStretch value: " + mScale);

				//修改开始.....................................................
				//图片等比例缩放，避免预览拉伸
                mScale = Math.min(
                        (canvas.getWidth() / (float)mCacheBitmap.getWidth()),
                        (canvas.getHeight() / (float)mCacheBitmap.getHeight())
                );

                if (mScale != 0) {
                    canvas.drawBitmap(mCacheBitmap, new Rect(0,0,mCacheBitmap.getWidth(), mCacheBitmap.getHeight()),
                         new Rect((int)((canvas.getWidth() - mScale*mCacheBitmap.getWidth())),
                         (int)((canvas.getHeight() - mScale*mCacheBitmap.getHeight())),
                         (int)((canvas.getWidth() - mScale*mCacheBitmap.getWidth()) + mScale*mCacheBitmap.getWidth()),
                         (int)((canvas.getHeight() - mScale*mCacheBitmap.getHeight()) + mScale*mCacheBitmap.getHeight())), null);
                } else {
                     canvas.drawBitmap(mCacheBitmap, new Rect(0,0,mCacheBitmap.getWidth(), mCacheBitmap.getHeight()),
                         new Rect((canvas.getWidth() - mCacheBitmap.getWidth()),
                         (canvas.getHeight() - mCacheBitmap.getHeight()),
                         (canvas.getWidth() - mCacheBitmap.getWidth()) + mCacheBitmap.getWidth(),
                         (canvas.getHeight() - mCacheBitmap.getHeight()) + mCacheBitmap.getHeight()), null);
                }
				//修改结束.....................................................
				
                if (mFpsMeter != null) {
                    mFpsMeter.measure();
                    mFpsMeter.draw(canvas, 20, 30);
                }
                getHolder().unlockCanvasAndPost(canvas);
            }
        }
    }

理解，整个流程应该分为以下几步：

1.手机通过摄像头捕捉到画面，并在摄像头监听接口的onCameraFrame里返回一个CvCameraViewFrame,其内包含两个Mat，此刻我们获取到了即将预览的图片对象mRgba

2.对该mRgba对象做旋转处理后，通过return mRgba给摄像头监听接口返回

3.此刻，在org.opencv.android.CameraBridgeViewBase文件里的deliverAndDrawFrame方法里，获取到了摄像头监听接口返回的该mRgba对象，即变量modified

		if (mListener != null) {
            modified = mListener.onCameraFrame(frame);
        } else {
            modified = frame.rgba();
        }

然后变量modified又转换成bitmap变量mCacheBitmap，而变量mCacheBitmap才是最终需要绘制到界面上的对象。
期间变量mCacheBitmap与变量modified所对应的图片长宽需要一致（需要修改AllocateCache方法），才能进行转换。

		if (modified != null) {
            try {
                Utils.matToBitmap(modified, mCacheBitmap);
            } catch(Exception e) {
                Log.e(TAG, "Mat type: " + modified);
                Log.e(TAG, "Bitmap type: " + mCacheBitmap.getWidth() + "*" + mCacheBitmap.getHeight());
                Log.e(TAG, "Utils.matToBitmap() throws an exception: " + e.getMessage());
                bmpValid = false;
            }
        }

转换好，获取到bitmap对象mCacheBitmap后，对其进行等比例缩放，即将mCacheBitmap缩放到画布对象canvas的大小，其中mScale 为缩放比例

		if (bmpValid && mCacheBitmap != null) {
            Canvas canvas = getHolder().lockCanvas();
            if (canvas != null) {
                canvas.drawColor(0, android.graphics.PorterDuff.Mode.CLEAR);
                if (BuildConfig.DEBUG)
                    Log.d(TAG, "mStretch value: " + mScale);

                mScale = Math.min(
                        (canvas.getWidth() / (float)mCacheBitmap.getWidth()),
                        (canvas.getHeight() / (float)mCacheBitmap.getHeight())
                );

                if (mScale != 0) {
                    canvas.drawBitmap(mCacheBitmap, new Rect(0,0,mCacheBitmap.getWidth(), mCacheBitmap.getHeight()),
                         new Rect((int)((canvas.getWidth() - mScale*mCacheBitmap.getWidth())),
                         (int)((canvas.getHeight() - mScale*mCacheBitmap.getHeight())),
                         (int)((canvas.getWidth() - mScale*mCacheBitmap.getWidth()) + mScale*mCacheBitmap.getWidth()),
                         (int)((canvas.getHeight() - mScale*mCacheBitmap.getHeight()) + mScale*mCacheBitmap.getHeight())), null);
                } else {
                     canvas.drawBitmap(mCacheBitmap, new Rect(0,0,mCacheBitmap.getWidth(), mCacheBitmap.getHeight()),
                         new Rect((canvas.getWidth() - mCacheBitmap.getWidth()),
                         (canvas.getHeight() - mCacheBitmap.getHeight()),
                         (canvas.getWidth() - mCacheBitmap.getWidth()) + mCacheBitmap.getWidth(),
                         (canvas.getHeight() - mCacheBitmap.getHeight()) + mCacheBitmap.getHeight()), null);
                }

                if (mFpsMeter != null) {
                    mFpsMeter.measure();
                    mFpsMeter.draw(canvas, 20, 30);
                }
                getHolder().unlockCanvasAndPost(canvas);
            }
        }

最终通过canvas.drawBitmap方法绘制出预览图片。
整个理解过程到此结束。
相较于网上较多的关于OpenCV横竖屏转换，比如

	//修改预览旋转90度问题
    canvas.rotate(90, 0, 0);
    float scalew = canvas.getWidth() / (float)mCacheBitmap.getHeight();
    float scaleh = canvas.getHeight() / (float)mCacheBitmap.getHeight();
    if (scaleh > scalew) {
        scalew = scaleh;
    }
    if (scalew !=0 ) {
        canvas.scale(scalew, scalew, 0, 0);
    }
    canvas.drawBitmap(mCacheBitmap, 0, -mCacheBitmap.getHeight(), null);
    //修改预览旋转90度问题end

旋转canvas引起的问题就是，如果你需要绘制文字，会发现文字方向被同步旋转了

Imgproc.putText(this.mRgba, "哥就是个传说",p3,
                            1 , 5, new Scalar(0,0,255),5,
                            4
                    );