This paper discusses how widespread use of computer vision's object identification algorithm has sparked substantial research into related fields including robotics, autonomous vehicles, scene interpretation, video surveillance, and more. Because of their centrality to these uses, visual recognition structures which include picture categorization, localization, and detection are receiving a lot of attention. The extraordinary progress made in neural networks, especially deep learning, has allowed for the visual recognition structures to reach a new level of performance. One area where computer vision has made remarkable progress is in object detection. This study serves as a metaphor for the significance of convolutional neural network–based deep learning algorithms for object identification. This work analyses current deep learning approaches to object identification. An example of a complete deep neural network for driverless cars is provided. In order to facilitate autonomous driving, we set out to create a deep neural network that could be used on embedded automotive systems. If you compare our proposed network architecture to the current state-of-the-art end-to-end deep neural networks used for autonomous driving, where the input to the machine learning algorithm is camera images and the output is the prediction of the steering angle, you'll see that ours is more straightforward

Deep learning, convolution neural network, object detection, visual recognition, computer vision