When i try to launch any of the apps (i tried edge_detection and find_contours) when specifying the image:="/camera/depth/image_rect_raw" the node crashes with console output below
process[edge_detection-1]: started with pid [16725]
terminate called after throwing an instance of 'cv_bridge::Exception'
what(): [16UC1] is not a color format. but [bgr8] is. The conversion does not make sense
[edge_detection-1] process has died [pid 16725, exit code -6, cmd /opt/ros/melodic/lib/opencv_apps/edge_detection image:=/camera/depth/image_rect_raw __name:=edge_detection __log:=/home/ahmad/.ros/log/1790e8f4-cf2d-11ea-8e04-54271e90ab2d/edge_detection-1.log].
log file: /home/ahmad/.ros/log/1790e8f4-cf2d-11ea-8e04-54271e90ab2d/edge_detection-1*.log
all processes on machine have died, roslaunch will exit
shutting down processing monitor...
... shutting down processing monitor complete
I understand the error is about the encoding used (bgr8). Is there a way to specify the depth map info (16uc1) as input?
Can this be edited to handle the depth map input? or is there a better/standard way?
Hi,
When i try to launch any of the apps (i tried edge_detection and find_contours) when specifying the image:="/camera/depth/image_rect_raw" the node crashes with console output below
I understand the error is about the encoding used (bgr8). Is there a way to specify the depth map info (16uc1) as input?
Can this be edited to handle the depth map input? or is there a better/standard way?
https://github.com/ros-perception/opencv_apps/blob/b3851233208b22bfe8d96fbc3ffa2fc06a00c5be/src/nodelet/edge_detection_nodelet.cpp#L139