It totally is: OpenCV does pixel-level image processing, in a deterministic and generally highly optimized way. For some computer vision, this is definitely eno

It totally is: OpenCV does pixel-level image processing, in a deterministic and generally highly optimized way. For some computer vision, this is definitely enough, and it will have a much better performance than doing it with neural networks. It might be complementary to higher-level frameworks (typically based on machine learning models), e.g. for object detection/classification.