Overview

This repository holds code of a ROS package for image acquisition with Black Fly camera. It is basically a ROS wrapper of the low-level API provided by Ponit Grey, the manufacturer of the camera. The ROS node can be configured to act as a server or as publisher at a given rate.

Dependencies

The package has been tested with the following dependencies:

• Ubuntu 16.04
• CMake + gcc
• ROS Kinetic
• OpenCV v3.1 or 3.2 (shipped with ROS Kinetic)
• Forked version of ROS common_msg metapackage. Clone it at your ROS workspace.
• FlyCapture2 (propietary SDK from the manufacturer)

To install the FlyCapture2 SDK dependency:

1. Download the SDK from the Point Grey's website
2. Install it by following the instructions in the README file included in the download.

Download and build this ROS package

Download to your ROS workspace /src, with the command:

$ git clone https://github.com/beta-robots/ptgrey_bfly_camera.git

and from your ROS workspace, build it with:

$ catkin_make --only-pkg-with-deps ptgrey_bfly_camera

ROS node configuration

Edit the .yaml file at config/ folder with the wished parameters. Specially set the run_mode, indicating if you want the camera to run as a SERVER (image as a service), or as a PUBLISHER (image streaming).

ROS node execution

Run the node (by default an rviz window will appear)

$ roslaunch ptgrey_bfly_camera ptgrey_bfly_camera.launch 

If the camera runs as a PUBLISHER, this node fulfills with the conventions specified here, about topics and service naming.

If the camera operates as a SERVER, please request the image service from another terminal, with the exposure parameter (0 means auto). This service will respond with a concatenation of two ROS sensor messages: Image and CameraInfo

$ rosservice call /ptgrey_bfly_camera/image_server "exposure: 0"

Intrinsics calibration and camera_info (topic and service)

The node will look for a yaml file, with the intrinsics calibration parameters, at folder /calibration. The camera_info topic will publish the parameters read from this file. If there is no file available, the node will run anyway, but the camera_info topic will publish invalid data. In case you need a calibrated camera, execute a procedure such as this one. At the end of such procedure, tha calibration node will call a service implemented in the camera node, requesting to store the calibration data, so a calibration file will be created and the camera_info topic will publish the new calibration parameters found.

Troubleshooting

The tool flycap, installed with camera's manufacturer SDK, has a set of options to configure the camera, like setting the device IP among others.

$ flycap