Cyber Traffic Light Visualizer

Requires:

python3
opencv-python
numpy

Installation:

If any changes are required to the .proto, protoc is also required.

Place traffic_light_pb2.py, and light_detection_project_p3.py into the following folder if there are any issues with running the scripts.

<$ROOT$>/apollo/cyber/python/cyber_py3/examples/

Place traffic_light_multi_cyberbag.py and traffic_light_pb2.py into the following folder if there are any issues with running the scripts.

<$ROOT$>/apollo/modules/tools/record_parse_save/

Usage:

For visualizing traffic light of a single cyberbag, run light_detection_project_p3.py.

• python light_detection_project_p3.py

This will create a short video of just the one file, as well as creating a visualizer with all the debug boxes from the traffic light topic.

---

For exporting videos of all traffic light intersections across multiple files, run traffic_light_multi_cyberbag.py.

NOTE: Update the export_folder variable to the desired export location, else the videos will be exported to the python script's location.

• python traffic_light_multi_cyberbag.py

This will read each file sequentially and create a new video for each individual intersection with a light, along with all the debug boxes from the traffic light topic.

For converting videos from .avi to .mp4, the following script is provided:

• avi_to_mp4.sh

Here is the usage:

Assumes current folder for both input, creates mp4_output folder in script directory for output.

• ./convert_avi_to_mp4.sh

Uses /path/to/input_folder for both input, creates mp4_output folder in script directory for output.

• ./convert_avi_to_mp4.sh /path/to/input_folder

Uses specified folders for input and output.

• ./convert_avi_to_mp4.sh /path/to/input_folder /path/to/output_folder