Threshold on Depth Image

[sdkaplan]

Required Info
Camera Model	D435i
Firmware Version	05.12.07.100
Operating System & Version	Ubuntu 18.04
Kernel Version (Linux Only)	4.9.140-tegra
Platform	Nvidia Jetson Xavier
SDK Version	2.35.2
Language	C++

I want to threshold the depth between two values without using realsense-ros. I currently have the code below. I am applying the filter to one depth map and not the other and then converting both images to opencv Mat so I can display using cv2::imshow(). The issue is that the threshold filter does not seem to be working. I am applying the filter and changing its parameters but nothing changes in the resulting image. Any suggestions? Am I using this filter wrong?

Code:
`int main(int argc, char **argv) try
{
// SET UP REALSENSE
rs2::colorizer color_map; // Declare depth colorizer for pretty visualization of depth
rs2::pipeline pipe; // Declare RealSense pipeline, encapsulating the device and sensors
// Start streaming with default recommended configuration. The default video configuration contains Depth and Color streams, Gyro and Accelerometer are enabled by default
pipe.start();

rs2::decimation_filter dec_filter;
rs2::spatial_filter spat_filter;
rs2::spatial_filter spat_filter_alpha;
rs2::threshold_filter thres_filter(0.15f, 4.f);
rs2::temporal_filter temp_filter;
 
rs2::frameset data = pipe.wait_for_frames();    // Wait for next set of frames from the camera
rs2::frame depth = data.get_depth_frame().apply_filter(color_map);
// Query frame size (width and height)
const int w = depth.as<rs2::video_frame>().get_width();
const int h = depth.as<rs2::video_frame>().get_height();

spat_filter_alpha.set_option(RS2_OPTION_FILTER_SMOOTH_ALPHA, 1);

spat_filter.set_option(RS2_OPTION_HOLES_FILL, 2);

while (waitKey(1) < 0){
    rs2::frameset data = pipe.wait_for_frames();// Wait for next set of frames
    rs2::frame depth = data.get_depth_frame().apply_filter(color_map); //rs depth frame
    rs2::frame depth1 = data.get_depth_frame().apply_filter(color_map); //rs depth frame
    depth = spat_filter_alpha.process(depth);
    depth = spat_filter.process(depth);    
    depth = thres_filter.process(depth);

    Mat frame_depth1(Size(w, h), CV_8UC3, (void*)depth1.get_data(), Mat::AUTO_STEP);
    Mat frame_depth(Size(w, h), CV_8UC3, (void*)depth.get_data(), Mat::AUTO_STEP);

    // Save frame to video
    imshow("depth filtered", frame_depth);
    imshow("depth raw", frame_depth1);

    //cv_ptr->image = frame_depth;
    //cv_ptr->encoding = "bgr8";
    //image_pub_.publish(cv_ptr->toImageMsg());

    // Exit if ESC pressed.
    int k = waitKey(1);
    if(k == 27){
        break;
    }

}

}
catch (const rs2::error & e)
{
std::cerr << "RealSense error calling " << e.get_failed_function() << "(" << e.get_failed_args() << "):\n " << e.what() << std::endl;
return EXIT_FAILURE;
}
catch (const std::exception& e)
{
std::cerr << e.what() << std::endl;
return EXIT_FAILURE;
}
`

[MartyG-RealSense]

Hi @sdkaplan Does the threshold filter work if you configure the minimum and maximum with the C++ instructions below, please?

thres_filter.set_option(RS2_OPTION_MIN_DISTANCE, 0.15);
thres_filter.set_option( RS2_OPTION_MAX_DISTANCE, 4.0);

[dorodnic]

I would also recommend applying threshold filter before colour map

[sdkaplan]

I applied both suggestions and it works now! Thanks so much!

[MartyG-RealSense]

That's great to hear @sdkaplan - thanks for the update!