The computer vision team is in charge of embedding artificial intelligence training models inside the vehicle's camera, which will help the submarine with object recognition and SLAM. Our main decision making mechanism is recognizing objects with machine learning through our cameras and processing that information to give desired outputs to logic algorithms.

​

To be able to complete the vision tasks specified in the competition, with the help of OpenCV, the algorithm needs to perform several image processing operations:

• Color detection and segmentation

• Shape detection

• Adaptive thresholding

• Contrast stretching

• Object segmentation

• Distance calculation from the object to the vehicle