Logging Visual Detections

Logging detections is done through the visual_logger node in the detection_logger package.

visual_logger will subscribe to the visual_detection topic, and reads detection msgs::ProcessedVisualDetection messages.

To get the end-to-end system running, follow these steps

1. Start roscore
2. Start the webcam feed with roslaunch video_stream_opencv webcam.launch
3. Start YOLO with roslaunch darknet_ros darknet_ros.launch
4. Start the image builder with rosrun image_draw image_draw_node
5. Start the logger with rosrun detection_logger visual_logger

The messages will get saved to a subfolder titled with the current year in the working directory.

specifics on outputted CSV file

To run the end-to-end system in Gazebo, follow these steps

go to[path to Detection-SARWAI]/src/darknet_ros/darknet_ros/config

and change in ros.yaml under the subscribers: section, change topic:to match the <imageTopicName> found in the model.sdf of the kinect model

Note: you will have to change this back to match the topic that any other cameras use to publish images

i.e. webcam.launch publishes to /webcam/image_raw

Start roscore

Start the webcam feed with roslaunch video_stream_opencv webcam.launch

Start YOLO with roslaunch darknet_ros darknet_ros.launch

Start the image builder with rosrun image_draw image_draw_node

Start the logger with rosrun detection_logger visual_logger