int main(int argc, char ** argv) {
// Keep track of the start time of the program
long long program_start_time = get_time();
// Let the user specify the number of frames to process
int num_frames = 1;
if (argc !=5){
fprintf(stderr, "usage: %s <num of frames> <num of threads><input file>", argv[0]);
exit(1);
}
if (argc > 1){
num_frames = atoi(argv[1]);
omp_num_threads = atoi(argv[2]);
omp_num_threads2 = atoi(argv[3]);
}
printf("Num of threads: %d\n", omp_num_threads);
printf("Num of threads: %d\n", omp_num_threads2);
omp_set_nested(1);
// Open video file
char *video_file_name;
video_file_name = argv[4];
avi_t *cell_file = AVI_open_input_file(video_file_name, 1);
if (cell_file == NULL) {
AVI_print_error("Error with AVI_open_input_file");
return -1;
}
int i, j, *crow, *ccol, pair_counter = 0, x_result_len = 0, Iter = 20, ns = 4, k_count = 0, n;
MAT *cellx, *celly, *A;
double *GICOV_spots, *t, *G, *x_result, *y_result, *V, *QAX_CENTERS, *QAY_CENTERS;
double threshold = 1.8, radius = 10.0, delta = 3.0, dt = 0.01, b = 5.0;
// Extract a cropped version of the first frame from the video file
MAT *image_chopped = get_frame(cell_file, 0, 1, 0);
printf("Detecting cells in frame 0\n");
// Get gradient matrices in x and y directions
MAT *grad_x = gradient_x(image_chopped);
MAT *grad_y = gradient_y(image_chopped);
m_free(image_chopped);
// Get GICOV matrix corresponding to image gradients
long long GICOV_start_time = get_time();
MAT *gicov = ellipsematching(grad_x, grad_y);
// Square GICOV values
MAT *max_gicov = m_get(gicov->m, gicov->n);
for (i = 0; i < gicov->m; i++) {
for (j = 0; j < gicov->n; j++) {
double val = m_get_val(gicov, i, j);
m_set_val(max_gicov, i, j, val * val);
}
}
long long GICOV_end_time = get_time();
// Dilate the GICOV matrix
long long dilate_start_time = get_time();
MAT *strel = structuring_element(12);
MAT *img_dilated = dilate_f(max_gicov, strel);
long long dilate_end_time = get_time();
// Find possible matches for cell centers based on GICOV and record the rows/columns in which they are found
pair_counter = 0;
crow = (int *) malloc(max_gicov->m * max_gicov->n * sizeof(int));
ccol = (int *) malloc(max_gicov->m * max_gicov->n * sizeof(int));
for (i = 0; i < max_gicov->m; i++) {
for (j = 0; j < max_gicov->n; j++) {
if (!(m_get_val(max_gicov,i,j) == 0.0) && (m_get_val(img_dilated,i,j) == m_get_val(max_gicov,i,j))) {
crow[pair_counter] = i;
ccol[pair_counter] = j;
pair_counter++;
}
}
}
GICOV_spots = (double *) malloc(sizeof(double)*pair_counter);
for (i = 0; i < pair_counter; i++)
GICOV_spots[i] = m_get_val(gicov, crow[i], ccol[i]);
G = (double *) calloc(pair_counter, sizeof(double));
x_result = (double *) calloc(pair_counter, sizeof(double));
y_result = (double *) calloc(pair_counter, sizeof(double));
x_result_len = 0;
for (i = 0; i < pair_counter; i++) {
if ((crow[i] > 29) && (crow[i] < BOTTOM - TOP + 39)) {
x_result[x_result_len] = ccol[i];
y_result[x_result_len] = crow[i] - 40;
G[x_result_len] = GICOV_spots[i];
x_result_len++;
}
}
// Make an array t which holds each "time step" for the possible cells
t = (double *) malloc(sizeof(double) * 36);
for (i = 0; i < 36; i++) {
t[i] = (double)i * 2.0 * PI / 36.0;
}
// Store cell boundaries (as simple circles) for all cells
cellx = m_get(x_result_len, 36);
celly = m_get(x_result_len, 36);
for(i = 0; i < x_result_len; i++) {
for(j = 0; j < 36; j++) {
m_set_val(cellx, i, j, x_result[i] + radius * cos(t[j]));
m_set_val(celly, i, j, y_result[i] + radius * sin(t[j]));
}
}
A = TMatrix(9,4);
V = (double *) malloc(sizeof(double) * pair_counter);
QAX_CENTERS = (double * )malloc(sizeof(double) * pair_counter);
QAY_CENTERS = (double *) malloc(sizeof(double) * pair_counter);
memset(V, 0, sizeof(double) * pair_counter);
memset(QAX_CENTERS, 0, sizeof(double) * pair_counter);
memset(QAY_CENTERS, 0, sizeof(double) * pair_counter);
// For all possible results, find the ones that are feasibly leukocytes and store their centers
k_count = 0;
for (n = 0; n < x_result_len; n++) {
if ((G[n] < -1 * threshold) || G[n] > threshold) {
MAT * x, *y;
VEC * x_row, * y_row;
x = m_get(1, 36);
y = m_get(1, 36);
x_row = v_get(36);
y_row = v_get(36);
// Get current values of possible cells from cellx/celly matrices
x_row = get_row(cellx, n, x_row);
y_row = get_row(celly, n, y_row);
uniformseg(x_row, y_row, x, y);
// Make sure that the possible leukocytes are not too close to the edge of the frame
if ((m_min(x) > b) && (m_min(y) > b) && (m_max(x) < cell_file->width - b) && (m_max(y) < cell_file->height - b)) {
MAT * Cx, * Cy, *Cy_temp, * Ix1, * Iy1;
VEC *Xs, *Ys, *W, *Nx, *Ny, *X, *Y;
Cx = m_get(1, 36);
Cy = m_get(1, 36);
Cx = mmtr_mlt(A, x, Cx);
Cy = mmtr_mlt(A, y, Cy);
Cy_temp = m_get(Cy->m, Cy->n);
for (i = 0; i < 9; i++)
m_set_val(Cy, i, 0, m_get_val(Cy, i, 0) + 40.0);
// Iteratively refine the snake/spline
for (i = 0; i < Iter; i++) {
int typeofcell;
if(G[n] > 0.0) typeofcell = 0;
else typeofcell = 1;
splineenergyform01(Cx, Cy, grad_x, grad_y, ns, delta, 2.0 * dt, typeofcell);
}
X = getsampling(Cx, ns);
for (i = 0; i < Cy->m; i++)
m_set_val(Cy_temp, i, 0, m_get_val(Cy, i, 0) - 40.0);
Y = getsampling(Cy_temp, ns);
Ix1 = linear_interp2(grad_x, X, Y);
Iy1 = linear_interp2(grad_x, X, Y);
Xs = getfdriv(Cx, ns);
Ys = getfdriv(Cy, ns);
Nx = v_get(Ys->dim);
for (i = 0; i < Ys->dim; i++)
v_set_val(Nx, i, v_get_val(Ys, i) / sqrt(v_get_val(Xs, i)*v_get_val(Xs, i) + v_get_val(Ys, i)*v_get_val(Ys, i)));
Ny = v_get(Xs->dim);
for (i = 0; i < Xs->dim; i++)
v_set_val(Ny, i, -1.0 * v_get_val(Xs, i) / sqrt(v_get_val(Xs, i)*v_get_val(Xs, i) + v_get_val(Ys, i)*v_get_val(Ys, i)));
W = v_get(Nx->dim);
for (i = 0; i < Nx->dim; i++)
v_set_val(W, i, m_get_val(Ix1, 0, i) * v_get_val(Nx, i) + m_get_val(Iy1, 0, i) * v_get_val(Ny, i));
V[n] = mean(W) / std_dev(W);
//get means of X and Y values for all "snaxels" of the spline contour, thus finding the cell centers
QAX_CENTERS[k_count] = mean(X);
QAY_CENTERS[k_count] = mean(Y) + TOP;
k_count++;
// Free memory
v_free(W);
v_free(Ny);
v_free(Nx);
v_free(Ys);
v_free(Xs);
m_free(Iy1);
m_free(Ix1);
v_free(Y);
v_free(X);
m_free(Cy_temp);
m_free(Cy);
m_free(Cx);
}
// Free memory
v_free(y_row);
v_free(x_row);
m_free(y);
m_free(x);
}
}
// Free memory
free(V);
free(ccol);
free(crow);
free(GICOV_spots);
free(t);
free(G);
free(x_result);
free(y_result);
m_free(A);
m_free(celly);
m_free(cellx);
m_free(img_dilated);
m_free(max_gicov);
m_free(gicov);
m_free(grad_y);
m_free(grad_x);
// Report the total number of cells detected
printf("Cells detected: %d\n\n", k_count);
// Report the breakdown of the detection runtime
printf("Detection runtime\n");
printf("-----------------\n");
printf("GICOV computation: %.5f seconds\n", ((float) (GICOV_end_time - GICOV_start_time)) / (1000*1000));
printf(" GICOV dilation: %.5f seconds\n", ((float) (dilate_end_time - dilate_start_time)) / (1000*1000));
printf(" Total: %.5f seconds\n", ((float) (get_time() - program_start_time)) / (1000*1000));
// Now that the cells have been detected in the first frame,
// track the ellipses through subsequent frames
if (num_frames > 1) printf("\nTracking cells across %d frames\n", num_frames);
else printf("\nTracking cells across 1 frame\n");
long long tracking_start_time = get_time();
int num_snaxels = 20;
ellipsetrack(cell_file, QAX_CENTERS, QAY_CENTERS, k_count, radius, num_snaxels, num_frames);
printf(" Total cambine: %.5f seconds\n", ((float) (get_time() - tracking_start_time)) / (float) (1000*1000*num_frames));
// Report total program execution time
printf("\nTotal application run time: %.5f seconds\n", ((float) (get_time() - program_start_time)) / (1000*1000));
return 0;
}
void VideoDemos( VideoCapture& surveillance_video, int starting_frame, bool clean_binary_images )
{
Mat previous_gray_frame, optical_flow, optical_flow_display;
Mat current_frame, thresholded_image, closed_image, first_frame;
Mat current_frame_gray, running_average_background;
Mat temp_running_average_background, running_average_difference;
Mat running_average_foreground_mask, running_average_foreground_image;
Mat selective_running_average_background;
Mat temp_selective_running_average_background, selective_running_average_difference;
Mat selective_running_average_foreground_mask, selective_running_average_background_mask, selective_running_average_foreground_image;
double running_average_learning_rate = 0.01;
surveillance_video.set(CV_CAP_PROP_POS_FRAMES,starting_frame);
surveillance_video >> current_frame;
first_frame = current_frame.clone();
cvtColor(current_frame, current_frame_gray, CV_BGR2GRAY);
current_frame.convertTo(running_average_background, CV_32F);
selective_running_average_background = running_average_background.clone();
int rad = running_average_background.depth();
MedianBackground median_background( current_frame, (float) 1.005, 1 );
Mat median_background_image, median_foreground_image;
int codec = static_cast<int>(surveillance_video.get(CV_CAP_PROP_FOURCC));
// V3.0.0 update on next line. OLD CODE was BackgroundSubtractorMOG2 gmm; //(50,16,true);
Ptr<BackgroundSubtractorMOG2> gmm = createBackgroundSubtractorMOG2();
Mat foreground_mask, foreground_image = Mat::zeros(current_frame.size(), CV_8UC3);
double frame_rate = surveillance_video.get(CV_CAP_PROP_FPS);
double time_between_frames = 1000.0/frame_rate;
Timestamper* timer = new Timestamper();
int frame_count = 0;
while ((!current_frame.empty()) && (frame_count++ < 1000))//1800))
{
double duration = static_cast<double>(getTickCount());
vector<Mat> input_planes(3);
split(current_frame,input_planes);
cvtColor(current_frame, current_frame_gray, CV_BGR2GRAY);
if (frame_count%2 == 0) // Skip every second frame so the flow is greater.
{
if ( previous_gray_frame.data )
{
Mat lucas_kanade_flow;
timer->ignoreTimeSinceLastRecorded();
LucasKanadeOpticalFlow(previous_gray_frame, current_frame_gray, lucas_kanade_flow);
timer->recordTime("Lucas Kanade Optical Flow");
calcOpticalFlowFarneback(previous_gray_frame, current_frame_gray, optical_flow, 0.5, 3, 15, 3, 5, 1.2, 0);
cvtColor(previous_gray_frame, optical_flow_display, CV_GRAY2BGR);
drawOpticalFlow(optical_flow, optical_flow_display, 8, Scalar(0, 255, 0), Scalar(0, 0, 255));
timer->recordTime("Farneback Optical Flow");
char frame_str[100];
sprintf( frame_str, "Frame = %d", frame_count);
Mat temp_output = JoinImagesHorizontally( current_frame, frame_str, optical_flow_display, "Farneback Optical Flow", 4 );
Mat optical_flow_output = JoinImagesHorizontally( temp_output, "", lucas_kanade_flow, "Lucas Kanade Optical Flow", 4 );
imshow("Optical Flow", optical_flow_output );
}
std::swap(previous_gray_frame, current_frame_gray);
}
// Static background image
Mat difference_frame, binary_difference;
Mat structuring_element(3,3,CV_8U,Scalar(1));
timer->ignoreTimeSinceLastRecorded();
absdiff(current_frame,first_frame,difference_frame);
cvtColor(difference_frame, thresholded_image, CV_BGR2GRAY);
threshold(thresholded_image,thresholded_image,30,255,THRESH_BINARY);
if (clean_binary_images)
{
morphologyEx(thresholded_image,closed_image,MORPH_CLOSE,structuring_element);
morphologyEx(closed_image,binary_difference,MORPH_OPEN,structuring_element);
current_frame.copyTo(binary_difference, thresholded_image);
}
else
{
binary_difference.setTo(Scalar(0,0,0));
current_frame.copyTo(binary_difference, thresholded_image);
}
timer->recordTime("Static difference");
// Running Average (three channel version)
vector<Mat> running_average_planes(3);
split(running_average_background,running_average_planes);
accumulateWeighted(input_planes[0], running_average_planes[0], running_average_learning_rate);
accumulateWeighted(input_planes[1], running_average_planes[1], running_average_learning_rate);
accumulateWeighted(input_planes[2], running_average_planes[2], running_average_learning_rate);
merge(running_average_planes,running_average_background);
running_average_background.convertTo(temp_running_average_background,CV_8U);
absdiff(temp_running_average_background,current_frame,running_average_difference);
split(running_average_difference,running_average_planes);
// Determine foreground points as any point with a difference of more than 30 on any one channel:
threshold(running_average_difference,running_average_foreground_mask,30,255,THRESH_BINARY);
split(running_average_foreground_mask,running_average_planes);
bitwise_or( running_average_planes[0], running_average_planes[1], running_average_foreground_mask );
bitwise_or( running_average_planes[2], running_average_foreground_mask, running_average_foreground_mask );
if (clean_binary_images)
{
morphologyEx(running_average_foreground_mask,closed_image,MORPH_CLOSE,structuring_element);
morphologyEx(closed_image,running_average_foreground_mask,MORPH_OPEN,structuring_element);
}
running_average_foreground_image.setTo(Scalar(0,0,0));
current_frame.copyTo(running_average_foreground_image, running_average_foreground_mask);
timer->recordTime("Running Average");
// Running Average with selective update
vector<Mat> selective_running_average_planes(3);
// Find Foreground mask
selective_running_average_background.convertTo(temp_selective_running_average_background,CV_8U);
absdiff(temp_selective_running_average_background,current_frame,selective_running_average_difference);
split(selective_running_average_difference,selective_running_average_planes);
// Determine foreground points as any point with an average difference of more than 30 over all channels:
Mat temp_sum = (selective_running_average_planes[0]/3 + selective_running_average_planes[1]/3 + selective_running_average_planes[2]/3);
threshold(temp_sum,selective_running_average_foreground_mask,30,255,THRESH_BINARY_INV);
// Update background
split(selective_running_average_background,selective_running_average_planes);
accumulateWeighted(input_planes[0], selective_running_average_planes[0], running_average_learning_rate,selective_running_average_foreground_mask);
accumulateWeighted(input_planes[1], selective_running_average_planes[1], running_average_learning_rate,selective_running_average_foreground_mask);
accumulateWeighted(input_planes[2], selective_running_average_planes[2], running_average_learning_rate,selective_running_average_foreground_mask);
invertImage(selective_running_average_foreground_mask,selective_running_average_foreground_mask);
accumulateWeighted(input_planes[0], selective_running_average_planes[0], running_average_learning_rate/3.0,selective_running_average_foreground_mask);
accumulateWeighted(input_planes[1], selective_running_average_planes[1], running_average_learning_rate/3.0,selective_running_average_foreground_mask);
accumulateWeighted(input_planes[2], selective_running_average_planes[2], running_average_learning_rate/3.0,selective_running_average_foreground_mask);
merge(selective_running_average_planes,selective_running_average_background);
if (clean_binary_images)
{
morphologyEx(selective_running_average_foreground_mask,closed_image,MORPH_CLOSE,structuring_element);
morphologyEx(closed_image,selective_running_average_foreground_mask,MORPH_OPEN,structuring_element);
}
selective_running_average_foreground_image.setTo(Scalar(0,0,0));
current_frame.copyTo(selective_running_average_foreground_image, selective_running_average_foreground_mask);
timer->recordTime("Selective Running Average");
// Median background
timer->ignoreTimeSinceLastRecorded();
median_background.UpdateBackground( current_frame );
timer->recordTime("Median");
median_background_image = median_background.GetBackgroundImage();
Mat median_difference;
absdiff(median_background_image,current_frame,median_difference);
cvtColor(median_difference, median_difference, CV_BGR2GRAY);
threshold(median_difference,median_difference,30,255,THRESH_BINARY);
median_foreground_image.setTo(Scalar(0,0,0));
current_frame.copyTo(median_foreground_image, median_difference);
// Update the Gaussian Mixture Model
// V3.0.0 update on next line. OLD CODE was gmm(current_frame, foreground_mask);
gmm->apply(current_frame, foreground_mask);
// Clean the resultant binary (moving pixel) mask using an opening.
threshold(foreground_mask,thresholded_image,150,255,THRESH_BINARY);
Mat moving_incl_shadows, shadow_points;
threshold(foreground_mask,moving_incl_shadows,50,255,THRESH_BINARY);
absdiff( thresholded_image, moving_incl_shadows, shadow_points );
Mat cleaned_foreground_mask;
if (clean_binary_images)
{
morphologyEx(thresholded_image,closed_image,MORPH_CLOSE,structuring_element);
morphologyEx(closed_image,cleaned_foreground_mask,MORPH_OPEN,structuring_element);
}
else cleaned_foreground_mask = thresholded_image.clone();
foreground_image.setTo(Scalar(0,0,0));
current_frame.copyTo(foreground_image, cleaned_foreground_mask);
timer->recordTime("Gaussian Mixture Model");
// Create an average background image (just for information)
Mat mean_background_image;
timer->ignoreTimeSinceLastRecorded();
// V3.0.0 update on next line. OLD CODE was gmm.getBackgroundImage(mean_background_image);
gmm->getBackgroundImage(mean_background_image);
duration = static_cast<double>(getTickCount())-duration;
duration /= getTickFrequency()/1000.0;
int delay = (time_between_frames>duration) ? ((int) (time_between_frames-duration)) : 1;
char c = cvWaitKey(delay);
char frame_str[100];
sprintf( frame_str, "Frame = %d", frame_count);
Mat temp_static_output = JoinImagesHorizontally( current_frame, frame_str, first_frame, "Static Background", 4 );
Mat static_output = JoinImagesHorizontally( temp_static_output, "", binary_difference, "Foreground", 4 );
imshow("Static Background Model", static_output );
Mat temp_running_output = JoinImagesHorizontally( current_frame, frame_str, temp_running_average_background, "Running Average Background", 4 );
Mat running_output = JoinImagesHorizontally( temp_running_output, "", running_average_foreground_image, "Foreground", 4 );
imshow("Running Average Background Model", running_output );
Mat temp_selective_output = JoinImagesHorizontally( current_frame, frame_str, temp_selective_running_average_background, "Selective Running Average Background", 4 );
Mat selective_output = JoinImagesHorizontally( temp_selective_output, "", selective_running_average_foreground_image, "Foreground", 4 );
imshow("Selective Running Average Background Model", selective_output );
Mat temp_median_output = JoinImagesHorizontally( current_frame, frame_str, median_background_image, "Median Background", 4 );
Mat median_output = JoinImagesHorizontally( temp_median_output, "", median_foreground_image, "Foreground", 4 );
imshow("Median Background Model", median_output );
Mat temp_gaussian_output = JoinImagesHorizontally( current_frame, frame_str, mean_background_image, "GMM Background", 4 );
Mat gaussian_output = JoinImagesHorizontally( temp_gaussian_output, "", foreground_image, "Foreground", 4 );
imshow("Gaussian Mixture Model", gaussian_output );
timer->putTimes( current_frame );
imshow( "Computation Times", current_frame );
surveillance_video >> current_frame;
}
cvDestroyAllWindows();
}
