Stereovision Algorithm Comparison

Project Description

This project compares different stereovision algorithms for disparity map estimation and real-time distance measurement. The implemented methods include:

• StereoBM
• StereoSGBM
• Hybrid-based - Cascade classifier combined with average color analysis of a selected area
• FastACVNet

This project was developed as part of my master's thesis.

Getting Started

1. Clone the repository

git clone https://github.com/sailor-elite/Stereovision_AlgorithmComparison.git
cd Stereovision_AlgorithmComparison

2. Install dependencies:

pip install -r requirements.txt

3. Run the main script:

python main.py

• In main.py, set the variables imgLeft and imgRight to the corresponding stereo image paths.
• Adjust parameters as needed.
• Adjust the IDs of your left and right cameras and run:

python Taking_Stereovision_Picture.py

• Adjust the IDs of your left and right cameras and run:

python Real_time_distance_measurement.py

Results of the hybrid based algorithm