Path Detection for Self-Driving Carts by using Canny Edge Detection Algorithm

A lane detection system is the most principal component of a self-driving transportation computation system. It deals with various complex scenarios such as lane curvature worn out marking, lane changing, turning U-turns, ending and separating lines. Many modern corporations and enterprises like Tesla have been able to achieve second grade i.e., partial automation at a commercial level. In this research paper, an improved and optimized lane detection method is proposed by using the OpenCV's canny edge detection algorithm. A quick and robust method that can easily detect lanes in a live video feed or a pre-recorded video stream. This proposed framework effectively combines the OpenCV's canny edge algorithm with the Hough transform function so it can be applied in small scale applications specifically for visitor carts in an apartment complex. Testing and analysis of the program shows that the suggested software is very reliable.