C++ InputArray::type方法代码示例

void cv::cuda::divide(InputArray _src1, InputArray _src2, OutputArray _dst, double scale, int dtype, Stream& stream)
{
    if (_src1.type() == CV_8UC4 && _src2.type() == CV_32FC1)
    {
        GpuMat src1 = _src1.getGpuMat();
        GpuMat src2 = _src2.getGpuMat();

        CV_Assert( src1.size() == src2.size() );

        _dst.create(src1.size(), src1.type());
        GpuMat dst = _dst.getGpuMat();

        divMat_8uc4_32f(src1, src2, dst, stream);
    }
    else if (_src1.type() == CV_16SC4 && _src2.type() == CV_32FC1)
    {
        GpuMat src1 = _src1.getGpuMat();
        GpuMat src2 = _src2.getGpuMat();

        CV_Assert( src1.size() == src2.size() );

        _dst.create(src1.size(), src1.type());
        GpuMat dst = _dst.getGpuMat();

        divMat_16sc4_32f(src1, src2, dst, stream);
    }
    else
    {
        arithm_op(_src1, _src2, _dst, GpuMat(), scale, dtype, stream, divMat, divScalar);
    }
}

void cv::viz::vtkTrajectorySource::SetTrajectory(InputArray _traj)
{
    CV_Assert(_traj.kind() == _InputArray::STD_VECTOR || _traj.kind() == _InputArray::MAT);
    CV_Assert(_traj.type() == CV_32FC(16) || _traj.type() == CV_64FC(16));

    Mat traj;
    _traj.getMat().convertTo(traj, CV_64F);
    const Affine3d* dpath = traj.ptr<Affine3d>();
    size_t total = traj.total();

    points = vtkSmartPointer<vtkPoints>::New();
    points->SetDataType(VTK_DOUBLE);
    points->SetNumberOfPoints((vtkIdType)total);

    tensors = vtkSmartPointer<vtkDoubleArray>::New();
    tensors->SetNumberOfComponents(9);
    tensors->SetNumberOfTuples((vtkIdType)total);

    for(size_t i = 0; i < total; ++i, ++dpath)
    {
        Matx33d R = dpath->rotation().t();  // transposed because of
        tensors->SetTuple((vtkIdType)i, R.val);        // column major order

        Vec3d p = dpath->translation();
        points->SetPoint((vtkIdType)i, p.val);
    }
}

double IPPE::PoseSolver::meanSceneDepth(InputArray _objectPoints, InputArray _rvec, InputArray _tvec)
{
    assert((_objectPoints.type() == CV_64FC3) | (_objectPoints.type() == CV_64FC3));

    size_t n = _objectPoints.rows() * _objectPoints.cols();
    Mat R;
    Mat q;
    Rodrigues(_rvec, R);
    double zBar = 0;

    for (size_t i = 0; i < n; i++) {
        cv::Mat p(_objectPoints.getMat().at<Point3d>(i));
        q = R * p + _tvec.getMat();
        double z;
        if (q.depth() == CV_64FC1) {
            z = q.at<double>(2);
        }
        else {
            z = static_cast<double>(q.at<float>(2));
        }
        zBar += z;

        //if (z <= 0) {
        //    std::cout << "Warning: object point " << i << " projects behind the camera! This should not be allowed. " << std::endl;
        //}
    }
    return zBar / static_cast<double>(n);
}

void cv::blendLinear( InputArray _src1, InputArray _src2, InputArray _weights1, InputArray _weights2, OutputArray _dst )
{
    int type = _src1.type(), depth = CV_MAT_DEPTH(type);
    Size size = _src1.size();

    CV_Assert(depth == CV_8U || depth == CV_32F);
    CV_Assert(size == _src2.size() && size == _weights1.size() && size == _weights2.size());
    CV_Assert(type == _src2.type() && _weights1.type() == CV_32FC1 && _weights2.type() == CV_32FC1);

    _dst.create(size, type);

    CV_OCL_RUN(_dst.isUMat(),
               ocl_blendLinear(_src1, _src2, _weights1, _weights2, _dst))

    Mat src1 = _src1.getMat(), src2 = _src2.getMat(), weights1 = _weights1.getMat(),
            weights2 = _weights2.getMat(), dst = _dst.getMat();

    if (depth == CV_8U)
    {
        BlendLinearInvoker<uchar> invoker(src1, src2, weights1, weights2, dst);
        parallel_for_(Range(0, src1.rows), invoker, dst.total()/(double)(1<<16));
    }
    else if (depth == CV_32F)
    {
        BlendLinearInvoker<float> invoker(src1, src2, weights1, weights2, dst);
        parallel_for_(Range(0, src1.rows), invoker, dst.total()/(double)(1<<16));
    }
}

void drawKeypoints( InputArray image, const std::vector<KeyPoint>& keypoints, InputOutputArray outImage,
                    const Scalar& _color, DrawMatchesFlags flags )
{
    CV_INSTRUMENT_REGION();

    if( !(flags & DrawMatchesFlags::DRAW_OVER_OUTIMG) )
    {
        if (image.type() == CV_8UC3 || image.type() == CV_8UC4)
        {
            image.copyTo(outImage);
        }
        else if( image.type() == CV_8UC1 )
        {
            cvtColor( image, outImage, COLOR_GRAY2BGR );
        }
        else
        {
            CV_Error( Error::StsBadArg, "Incorrect type of input image: " + typeToString(image.type()) );
        }
    }

    RNG& rng=theRNG();
    bool isRandColor = _color == Scalar::all(-1);

    CV_Assert( !outImage.empty() );
    std::vector<KeyPoint>::const_iterator it = keypoints.begin(),
                                     end = keypoints.end();
    for( ; it != end; ++it )
    {
        Scalar color = isRandColor ? Scalar( rng(256), rng(256), rng(256), 255 ) : _color;
        _drawKeypoint( outImage, *it, color, flags );
    }
}

void cv::cuda::multiply(InputArray _src1, InputArray _src2, OutputArray _dst, double scale, int dtype, Stream& stream)
{
    if (_src1.type() == CV_8UC4 && _src2.type() == CV_32FC1)
    {
        GpuMat src1 = getInputMat(_src1, stream);
        GpuMat src2 = getInputMat(_src2, stream);

        CV_Assert( src1.size() == src2.size() );

        GpuMat dst = getOutputMat(_dst, src1.size(), src1.type(), stream);

        mulMat_8uc4_32f(src1, src2, dst, stream);

        syncOutput(dst, _dst, stream);
    }
    else if (_src1.type() == CV_16SC4 && _src2.type() == CV_32FC1)
    {
        GpuMat src1 = getInputMat(_src1, stream);
        GpuMat src2 = getInputMat(_src2, stream);

        CV_Assert( src1.size() == src2.size() );

        GpuMat dst = getOutputMat(_dst, src1.size(), src1.type(), stream);

        mulMat_16sc4_32f(src1, src2, dst, stream);

        syncOutput(dst, _dst, stream);
    }
    else
    {
        arithm_op(_src1, _src2, _dst, GpuMat(), scale, dtype, stream, mulMat, mulScalar);
    }
}

int matte(InputArray src, OutputArray dst, Point firstPoint, Point secondPoint, float sigmaX,
            float sigmaY, Size ksize)
{
    CV_Assert((src.type() == CV_8UC3) || (src.type() == CV_32FC3));
    CV_Assert((sigmaX > 0.0f) || (sigmaY > 0.0f));
    Mat srcImg = src.getMat();
    CV_Assert(!(srcImg.empty()));
    if(srcImg.type() != CV_32FC3)
    {
        srcImg.convertTo(srcImg, CV_32FC3, 1.0f/255.0f);
    }
    int xsize = firstPoint.x - secondPoint.x;
    int ysize = firstPoint.y - secondPoint.y;
    Point topLeft = topleftFind(firstPoint, secondPoint, xsize, ysize);
    const Scalar black = Scalar(0.0f,0.0f,0.0f);
    const Scalar white = Scalar(1.0f,1.0f,1.0f);
    Mat mask(srcImg.rows, srcImg.cols, CV_32FC1, black);
    ellipse(mask, Point((topLeft.x+xsize/2),(topLeft.y-ysize/2)),
            Size(xsize/2,ysize/2), 0, 0, 360, white, -1);
    GaussianBlur(mask, mask, ksize, sigmaX, sigmaY);
    vector<Mat> ch_img;
    split(srcImg,ch_img);
    ch_img[0]=ch_img[0].mul(mask)+1.0f-mask;
    ch_img[1]=ch_img[1].mul(mask)+1.0f-mask;
    ch_img[2]=ch_img[2].mul(mask)+1.0f-mask;
    merge(ch_img,dst);
    return 0;
}

void IPPE::PoseSolver::homographyFromSquarePoints(InputArray _targetPoints, double halfLength, OutputArray H_)
{

    assert((_targetPoints.type() == CV_32FC2) | (_targetPoints.type() == CV_64FC2));

    cv::Mat pts = _targetPoints.getMat();
    H_.create(3, 3, CV_64FC1);
    Mat H = H_.getMat();

    double p1x, p1y;
    double p2x, p2y;
    double p3x, p3y;
    double p4x, p4y;

    if (_targetPoints.type() == CV_32FC2) {
        p1x = -pts.at<Vec2f>(0)(0);
        p1y = -pts.at<Vec2f>(0)(1);

        p2x = -pts.at<Vec2f>(1)(0);
        p2y = -pts.at<Vec2f>(1)(1);

        p3x = -pts.at<Vec2f>(2)(0);
        p3y = -pts.at<Vec2f>(2)(1);

        p4x = -pts.at<Vec2f>(3)(0);
        p4y = -pts.at<Vec2f>(3)(1);
    }
    else {
        p1x = -pts.at<Vec2d>(0)(0);
        p1y = -pts.at<Vec2d>(0)(1);

        p2x = -pts.at<Vec2d>(1)(0);
        p2y = -pts.at<Vec2d>(1)(1);

        p3x = -pts.at<Vec2d>(2)(0);
        p3y = -pts.at<Vec2d>(2)(1);

        p4x = -pts.at<Vec2d>(3)(0);
        p4y = -pts.at<Vec2d>(3)(1);
    }

    //analytic solution:
    double detsInv = -1 / (halfLength * (p1x * p2y - p2x * p1y - p1x * p4y + p2x * p3y - p3x * p2y + p4x * p1y + p3x * p4y - p4x * p3y));

    H.at<double>(0, 0) = detsInv * (p1x * p3x * p2y - p2x * p3x * p1y - p1x * p4x * p2y + p2x * p4x * p1y - p1x * p3x * p4y + p1x * p4x * p3y + p2x * p3x * p4y - p2x * p4x * p3y);
    H.at<double>(0, 1) = detsInv * (p1x * p2x * p3y - p1x * p3x * p2y - p1x * p2x * p4y + p2x * p4x * p1y + p1x * p3x * p4y - p3x * p4x * p1y - p2x * p4x * p3y + p3x * p4x * p2y);
    H.at<double>(0, 2) = detsInv * halfLength * (p1x * p2x * p3y - p2x * p3x * p1y - p1x * p2x * p4y + p1x * p4x * p2y - p1x * p4x * p3y + p3x * p4x * p1y + p2x * p3x * p4y - p3x * p4x * p2y);
    H.at<double>(1, 0) = detsInv * (p1x * p2y * p3y - p2x * p1y * p3y - p1x * p2y * p4y + p2x * p1y * p4y - p3x * p1y * p4y + p4x * p1y * p3y + p3x * p2y * p4y - p4x * p2y * p3y);
    H.at<double>(1, 1) = detsInv * (p2x * p1y * p3y - p3x * p1y * p2y - p1x * p2y * p4y + p4x * p1y * p2y + p1x * p3y * p4y - p4x * p1y * p3y - p2x * p3y * p4y + p3x * p2y * p4y);
    H.at<double>(1, 2) = detsInv * halfLength * (p1x * p2y * p3y - p3x * p1y * p2y - p2x * p1y * p4y + p4x * p1y * p2y - p1x * p3y * p4y + p3x * p1y * p4y + p2x * p3y * p4y - p4x * p2y * p3y);
    H.at<double>(2, 0) = -detsInv * (p1x * p3y - p3x * p1y - p1x * p4y - p2x * p3y + p3x * p2y + p4x * p1y + p2x * p4y - p4x * p2y);
    H.at<double>(2, 1) = detsInv * (p1x * p2y - p2x * p1y - p1x * p3y + p3x * p1y + p2x * p4y - p4x * p2y - p3x * p4y + p4x * p3y);
    H.at<double>(2, 2) = 1.0;
}

void NullOutlierRejector::process(
        Size /*frameSize*/, InputArray points0, InputArray points1, OutputArray mask)
{
    CV_Assert(points0.type() == points1.type());
    CV_Assert(points0.getMat().checkVector(2) == points1.getMat().checkVector(2));

    int npoints = points0.getMat().checkVector(2);
    mask.create(1, npoints, CV_8U);
    Mat mask_ = mask.getMat();
    mask_.setTo(1);
}

void addNoise(InputArray src_, OutputArray dest_, const double sigma, const double sprate, const int seed)
{
	if(seed!=0) cv::theRNG().state = seed;
	if (dest_.empty() || dest_.size() != src_.size() || dest_.type() != src_.type()) dest_.create(src_.size(), src_.type());
	Mat src = src_.getMat();
	Mat dest = dest_.getMat();
	if (src.channels() == 1)
	{
		addNoiseMono(src, dest, sigma);
		if (sprate != 0)addNoiseSoltPepperMono(dest, dest, sprate, seed);
		return;
	}
	else
	{
		vector<Mat> s(src.channels());
		vector<Mat> d(src.channels());
		split(src, s);
		for (int i = 0; i < src.channels(); i++)
		{
			addNoiseMono(s[i], d[i], sigma);
			if (sprate != 0)addNoiseSoltPepperMono(d[i], d[i], sprate, seed);
		}
		cv::merge(d, dest);
	}
	if (seed != 0) cv::theRNG().state = cv::getTickCount();
}

static void elbp(InputArray src, OutputArray dst, int radius, int neighbors)
{
    int type = src.type();

    switch (type)
    {
        case CV_8SC1:   elbp_<char>(src,dst, radius, neighbors);
            break;
        case CV_8UC1:   elbp_<unsigned char>(src, dst, radius, neighbors);
            break;
        case CV_16SC1:  elbp_<short>(src,dst, radius, neighbors);
            break;
        case CV_16UC1:  elbp_<unsigned short>(src,dst, radius, neighbors);
            break;
        case CV_32SC1:  elbp_<int>(src,dst, radius, neighbors);
            break;
        case CV_32FC1:  elbp_<float>(src,dst, radius, neighbors);
            break;
        case CV_64FC1:  elbp_<double>(src,dst, radius, neighbors);
            break;
        default:
            String error_msg = format("Using Original Local Binary Patterns for feature extraction only works on single-channel images (given %d). Please pass the image data as a grayscale image!", type);
            CV_Error(CV_StsNotImplemented, error_msg);
            break;
    }
}

/*!
Niblack binarization algorithm.

@param src [in] Mat, single channel uchar image.
@param dst [out] Mat, result image.
@param windowSize [in] int, window size for calculation.
@param k [in] int, parameter for local threshold.
@return int, 0x0000 = Success.
*/
int Niblack(InputArray src, OutputArray dst, int windowSize, float k)
{
	if (src.type() != CV_8UC1 || src.empty())	
		return 0x0001;	/*!< source image type not supported. */

	/*! update window size, which should be odd. */
	if (windowSize < 2)	
		return 0x0002;	/*!< window size not supported. */
	if (windowSize / 2 == 0)
		windowSize++;

	Mat source, destination;
	Mat sourceUchar = src.getMat();
	sourceUchar.convertTo(source, CV_32FC1);

	/*! calcalte mean and variance via
	D(x) = E(x^2) - (Ex)^2 */
	Mat avg, power, avg_power, power_avg;
	Mat standard;
	boxFilter(source, avg, -1, Size(windowSize, windowSize));
	pow(avg, 2, avg_power);
	pow(source, 2, power);
	boxFilter(power, power_avg, -1, Size(windowSize, windowSize));
	sqrt(power_avg - power_avg, standard);

	/*! calculate local threshold */
	Mat threshold = avg + k * standard;

	/*! Output result */
	dst.create(sourceUchar.size(), CV_8UC1);
	destination = dst.getMat();
	destination = source > threshold;

	return 0x0000;
}

void repeat(InputArray _src, int ny, int nx, OutputArray _dst)
{
    CV_Assert( _src.dims() <= 2 );
    CV_Assert( ny > 0 && nx > 0 );

    Size ssize = _src.size();
    _dst.create(ssize.height*ny, ssize.width*nx, _src.type());

    CV_OCL_RUN(_dst.isUMat(),
               ocl_repeat(_src, ny, nx, _dst))

    Mat src = _src.getMat(), dst = _dst.getMat();
    Size dsize = dst.size();
    int esz = (int)src.elemSize();
    int x, y;
    ssize.width *= esz; dsize.width *= esz;

    for( y = 0; y < ssize.height; y++ )
    {
        for( x = 0; x < dsize.width; x += ssize.width )
            memcpy( dst.data + y*dst.step + x, src.data + y*src.step, ssize.width );
    }

    for( ; y < dsize.height; y++ )
        memcpy( dst.data + y*dst.step, dst.data + (y - ssize.height)*dst.step, dsize.width );
}

    void detect( InputArray _image, std::vector<KeyPoint>& keypoints, InputArray _mask )
    {
        CV_INSTRUMENT_REGION()

        std::vector<Point2f> corners;

        if (_image.isUMat())
        {
            UMat ugrayImage;
            if( _image.type() != CV_8U )
                cvtColor( _image, ugrayImage, COLOR_BGR2GRAY );
            else
                ugrayImage = _image.getUMat();

            goodFeaturesToTrack( ugrayImage, corners, nfeatures, qualityLevel, minDistance, _mask,
                                 blockSize, useHarrisDetector, k );
        }
        else
        {
            Mat image = _image.getMat(), grayImage = image;
            if( image.type() != CV_8U )
                cvtColor( image, grayImage, COLOR_BGR2GRAY );

            goodFeaturesToTrack( grayImage, corners, nfeatures, qualityLevel, minDistance, _mask,
                                blockSize, useHarrisDetector, k );
        }

        keypoints.resize(corners.size());
        std::vector<Point2f>::const_iterator corner_it = corners.begin();
        std::vector<KeyPoint>::iterator keypoint_it = keypoints.begin();
        for( ; corner_it != corners.end(); ++corner_it, ++keypoint_it )
            *keypoint_it = KeyPoint( *corner_it, (float)blockSize );

    }

static bool ocl_integral( InputArray _src, OutputArray _sum, int sdepth )
{
    if ( _src.type() != CV_8UC1 || _src.step() % vlen != 0 || _src.offset() % vlen != 0  ||
         !(sdepth == CV_32S || sdepth == CV_32F) )
        return false;

    ocl::Kernel k1("integral_sum_cols", ocl::imgproc::integral_sum_oclsrc,
                   format("-D sdepth=%d", sdepth));
    if (k1.empty())
        return false;

    Size size = _src.size(), t_size = Size(((size.height + vlen - 1) / vlen) * vlen, size.width),
            ssize(size.width + 1, size.height + 1);
    _sum.create(ssize, sdepth);
    UMat src = _src.getUMat(), t_sum(t_size, sdepth), sum = _sum.getUMat();
    t_sum = t_sum(Range::all(), Range(0, size.height));

    int offset = (int)src.offset / vlen, pre_invalid = (int)src.offset % vlen;
    int vcols = (pre_invalid + src.cols + vlen - 1) / vlen;
    int sum_offset = (int)sum.offset / vlen;

    k1.args(ocl::KernelArg::PtrReadOnly(src), ocl::KernelArg::PtrWriteOnly(t_sum),
            offset, pre_invalid, src.rows, src.cols, (int)src.step, (int)t_sum.step);
    size_t gt = ((vcols + 1) / 2) * 256, lt = 256;
    if (!k1.run(1, &gt, &lt, false))
        return false;

    ocl::Kernel k2("integral_sum_rows", ocl::imgproc::integral_sum_oclsrc,
                   format("-D sdepth=%d", sdepth));
    k2.args(ocl::KernelArg::PtrReadWrite(t_sum), ocl::KernelArg::PtrWriteOnly(sum),
            t_sum.rows, t_sum.cols, (int)t_sum.step, (int)sum.step, sum_offset);

    size_t gt2 = t_sum.cols  * 32, lt2 = 256;
    return k2.run(1, &gt2, &lt2, false);
}