I try to calculate a SIFT Descriptor out of KeyPoints found by the FAST Corner Detection. For that, I adapted some functions from the Sift.cpp and changed them so they can be used with FAST generated KeyPoints with nonmaxsuppression. While I try to calculate the Descriptor I get an EXC_BAD_ACCESS Error and I cant find the origin of this Error. I tried to make a example as small as possible of my code where the Error is generated with no constants or aliases. When you want to know more about the full code see my Git or look at this Question where already some things have been explained in the comments. Now the code:
int descriptorSize() {
// Descr_Width * Descr_Width * Descr_hist_bins
return 3*3*4;
}
static void calcSiftDescriptor( const Mat& img, Point2f ptf, float ori, float scl,
int d, int n, float* dst )
{
Point pt(cvRound(ptf.x), cvRound(ptf.y));
float cos_t = cosf(ori*(float)(CV_PI/180));
float sin_t = sinf(ori*(float)(CV_PI/180));
float bins_per_rad = n / 360.f;
float exp_scale = -1.f/(d * d * 0.5f);
// 3.f = Descr_scale_factor
float hist_width = 3.f * scl;
int radius = cvRound(hist_width * 1.4142135623730951f * (d + 1) * 0.5f);
// Clip the radius to the diagonal of the image to avoid autobuffer too large exception
radius = std::min(radius, (int) sqrt(((double) img.cols)*img.cols + ((double) img.rows)*img.rows));
cos_t /= hist_width;
sin_t /= hist_width;
int i, j, k, len = (radius*2+1)*(radius*2+1), histlen = (d+2)*(d+2)*(n+2);
int rows = img.rows, cols = img.cols;
AutoBuffer<float> buf(len*6 + histlen);
float *X = buf, *Y = X + len, *Mag = Y, *Ori = Mag + len, *W = Ori + len;
float *RBin = W + len, *CBin = RBin + len, *hist = CBin + len;
for( i = 0; i < d+2; i++ )
{
for( j = 0; j < d+2; j++ )
for( k = 0; k < n+2; k++ )
hist[(i*(d+2) + j)*(n+2) + k] = 0.;
}
for( i = -radius, k = 0; i <= radius; i++ )
for( j = -radius; j <= radius; j++ )
{
// Calculate sample's histogram array coords rotated relative to ori.
// Subtract 0.5 so samples that fall e.g. in the center of row 1 (i.e.
// r_rot = 1.5) have full weight placed in row 1 after interpolation.
float c_rot = j * cos_t - i * sin_t;
float r_rot = j * sin_t + i * cos_t;
float rbin = r_rot + d/2 - 0.5f;
float cbin = c_rot + d/2 - 0.5f;
int r = pt.y + i, c = pt.x + j;
if( rbin > -1 && rbin < d && cbin > -1 && cbin < d &&
r > 0 && r < rows - 1 && c > 0 && c < cols - 1 )
{
float dx = (float)(img.at<float>(r, c+1) - img.at<float>(r, c-1));
float dy = (float)(img.at<float>(r-1, c) - img.at<float>(r+1, c));
X[k] = dx; Y[k] = dy; RBin[k] = rbin; CBin[k] = cbin;
W[k] = (c_rot * c_rot + r_rot * r_rot)*exp_scale;
k++;
}
}
len = k;
cv::hal::fastAtan2(Y, X, Ori, len, true);
cv::hal::magnitude32f(X, Y, Mag, len);
cv::hal::exp32f(W, W, len);
for( k = 0; k < len; k++ )
{
float rbin = RBin[k], cbin = CBin[k];
float obin = (Ori[k] - ori)*bins_per_rad;
float mag = Mag[k]*W[k];
int r0 = cvFloor( rbin );
int c0 = cvFloor( cbin );
int o0 = cvFloor( obin );
rbin -= r0;
cbin -= c0;
obin -= o0;
if( o0 < 0 )
o0 += n;
if( o0 >= n )
o0 -= n;
// histogram update using tri-linear interpolation
float v_r1 = mag*rbin, v_r0 = mag - v_r1;
float v_rc11 = v_r1*cbin, v_rc10 = v_r1 - v_rc11;
float v_rc01 = v_r0*cbin, v_rc00 = v_r0 - v_rc01;
float v_rco111 = v_rc11*obin, v_rco110 = v_rc11 - v_rco111;
float v_rco101 = v_rc10*obin, v_rco100 = v_rc10 - v_rco101;
float v_rco011 = v_rc01*obin, v_rco010 = v_rc01 - v_rco011;
float v_rco001 = v_rc00*obin, v_rco000 = v_rc00 - v_rco001;
int idx = ((r0+1)*(d+2) + c0+1)*(n+2) + o0;
hist[idx] += v_rco000; //EXC_BAD_ACCESS here
hist[idx+1] += v_rco001;
hist[idx+(n+2)] += v_rco010;
hist[idx+(n+3)] += v_rco011;
hist[idx+(d+2)*(n+2)] += v_rco100;
hist[idx+(d+2)*(n+2)+1] += v_rco101;
hist[idx+(d+3)*(n+2)] += v_rco110;
hist[idx+(d+3)*(n+2)+1] += v_rco111;
}
// finalize histogram, since the orientation histograms are circular
for( i = 0; i < d; i++ )
for( j = 0; j < d; j++ )
{
int idx = ((i+1)*(d+2) + (j+1))*(n+2);
hist[idx] += hist[idx+n];
hist[idx+1] += hist[idx+n+1];
for( k = 0; k < n; k++ )
dst[(i*d + j)*n + k] = hist[idx+k];
}
// copy histogram to the descriptor,
// apply hysteresis thresholding
// and scale the result, so that it can be easily converted
// to byte array
float nrm2 = 0;
len = d*d*n;
for( k = 0; k < len; k++ )
nrm2 += dst[k]*dst[k];
//0.2f = Descriptor Mag Threshold
float thr = std::sqrt(nrm2)*0.2f;
for( i = 0, nrm2 = 0; i < k; i++ )
{
float val = std::min(dst[i], thr);
dst[i] = val;
nrm2 += val*val;
}
//512.f = Factor to convert floating point descriptor to uchar
nrm2 = 512.f/std::max(std::sqrt(nrm2), FLT_EPSILON);
for( k = 0; k < len; k++ )
{
dst[k] = saturate_cast<uchar>(dst[k]*nrm2);
}
}
Mat src, gry, dst, tmp;
int main(int argc, const char * argv[]) {
Point2f pt(351, 119);
float ori = 2.97183228;
String imgName("/users/mtuchner/desktop/test/test/example.jpg");
src = imread(imgName);
cvtColor(src, gry, CV_BGR2GRAY);
int rows = gry.rows/sqrt(2);
int cols = gry.cols/sqrt(2);
resize(gry, tmp, Size(rows, cols), 0, 0, INTER_NEAREST);
GaussianBlur(tmp, dst, Size(5,5), 0, 0);
//Detection of keypoints here. Normally in a Loop where the Image is always made
//smaler until the Size of the found Features is < 100.
int dsize = descriptorSize();
// 489 = Number of all Features found in the Detection.
Mat descriptors(489, dsize, CV_32F);
int i = 3;
// 3.5 is the size normally calculated from the keyPoint size,
// 3 is the Descriptor width and 4 the Bins in the Descriptor Histogram
calcSiftDescriptor(dst, pt, ori, 3.5, 3, 4, descriptors.ptr<float>((int)i));
return 0;
}
The Value of Idx is in this case totally strange. (-2147483648) The Origin could be the Orientation Matrix Ori which results out of the function hal::fastAtan2 the value of Y[k] = NaN which normally shouldn't be because the corresponding grab for dyis still in the Image and not out of bounds.
