|
|
VisionWorks Toolkit ReferenceApril 8, 2016 | 1.4 Release |
Hough Transform Demo user guide.
nvx_demo_hough_transform demonstrates lines and circles detection via Hough Transform.
The demo uses the following pipeline:
(frame)
|
[ColorConvert]
|
[ChannelExtract]
|
[ScaleImage (down)]
|
[Median3x3]
|
[EqualizeHist]
|
+----------------------+----------------------+
| |
[CannyEdgeDetector] |
| |
+---+---------------------+----+ [Sobel3x3]
| | | |
| | +------------------+
| | |
[ScaleImage (up)] [HoughSegments] [HoughCircles]
| | |
(edges) (segments) (circles)
The input frame is converted to grayscale, downscaled, blurred with Median filter, and equalized. The equalized frame is then processed by Canny Edge Detector and Sobel operator, and the resulting edges image and derivatives are passed to the Hough Circle node to get the final array with detected circles. The edges image is also passed to the Hough Segments node to get the final array with detected lines.
nvx_demo_hough_transform is installed in the following directory:
/usr/share/visionworks/sources/demos/hough_transform
For the steps to build sample applications, see the Samples and Demos section for your OS.
./nvx_demo_hough_transform [options]
This topic provides a list of supported options and the values they consume.
--source=/path/to/video.avi for video--source=/path/to/image for image--source=/path/to/image_%04d_sequence.png for image sequence--source="device:///v4l2?index=0" for the first V4L2 camera--source="device:///v4l2?index=1" for the second V4L2 camera.--source="device:///nvcamera?index=0" for the GStreamer NVIDIA camera (Jetson TX1 only).--source="device:///nvmedia?config=dvp-ov10635-yuv422-ab-e2379&number=4" for the GStreamer NVIDIA camera (Vibrante for Linux only).sudo.Description: Specifies the path to the configuration file.
This file contains the following parameters:
Usage:
./nvx_demo_hough_transform --config=/path/to/config_file.ini
M to switch source/edges.Space to pause/resume the demo.ESC to close the demo.