Vision Based Intelligent Vehicle Control /
Savera Yousaf, Zoha Ali Kayani, Muhammad Talha Khalid, Hassaan Bin Nadeem. (TCC-31 / BETE-56)
- 64 p
This project develops prototype of self-driving car. The vehicle is equipped with camera. To easily identify road, we used histogram analysis. This method requires little memory and lower processing power. A raspberry pi is sufficient for it. For preprocessing, we applied binary thresholding, perspective warping and region of interest. The car can drive itself on the road. Drive optimization is achieved by averaging method for smoothing the road. In this novel method, use of histogram makes the process applicable in small computers. As small computers like Raspberry Pi have very small amount of computing capability, our system is capable to run the full process. The image processing method we proposed here is far better for efficient use of memory and processing power. We used histogram for its superiority in any color intensity variation. This aim of this project is to develop an autonomous vehicle that can efficiently and safely navigate roads by detecting the curve direction without human intervention. We integrate advanced technologies like computer vision, image processing, and machine learning to enable the vehicle to interpret and perceive its environment to make decisions and control the movements of the car based on its decisions.