Daheer/Driving-Environment-Detector

Detecting road objects using YOLO CNN Architecture

Driving-Environment-Detector

Driving-Environment-Detector recognizes everyday road objects on a road scene. It is based on the You Only Look Once CNN Architecture, specifically the YOLO v2 Darknet 19.

Model Architecture Plot

Click to view full architecture

Built Using

• Python
• Tensorflow
• OpenCV
• Others

Prerequisites and Installation

Python

python driving_environment_detector.py

Project Structure

.
├── README.md
├── FiraMono-Medium.otf
├── SIL Open Font License.txt
├── Images
│   ├── sample.png
│   ├── yolo_model_architecture_short.png
│   ├── yolo_model_architecture.png
│   ├── yolo v2 darknet19.png
│   ├── sample_input.png
│   └── sample_input.png
├── model data
│   ├── variables
│   │   ├── anchors.txt
│   │   ├── coco_classes.txt
│   │   ├── pascal_classes.txt
│   │   ├── saved_model.pb
│   │   └── yolo_anchors.txt
│   └── yad2k
│   │   ├── __pycache__
│   │   ├── models
│   │   └── utils
│   │   │   └── util.py
├── .gitattributes
├── driving_environment_detector_voila.ipynb
├── driving_environment_detector.ipynb
├── driving_environment_detector.py
└── requirements.txt

Usage

Simply place your video covering a road scene in the top directory. Run the installation code, sip some coffee or take a walk depending on the legth of your video :). When completed, the new video can be found in out/output_video.mp4

Demo

Sample Input	Sample Output

References

• SlimDeblurGAN-Based Motion Deblurring and Marker Detection for Autonomous Drone Landing - Scientific Figure on ResearchGate.
• Convolutional Neural Netwokrs

Contact

Dahir Ibrahim (Deedax Inc) - http://instagram.com/deedax_inc
Email - suhayrid@gmail.com
YouTube - https://www.youtube.com/channel/UCqvDiAJr2gRREn2tVtXFhvQ
Project Link - https://github.com/Daheer/Driving-Environment-Detector
Twitter - https://twitter.com/DeedaxInc