• The main objective of my project : to recognize objects of multiple classes in images and videos, I used machine learning algorithms and deep learning (MLP, CNN, RNN ...), which have led to a great revolution in the field of artificial intelligence. In particular they have been successfully used for object classification in images and videos. • Working environment : Anaconda, PyCharm, Google Colaboratory, LATEX (Report). • Languages and libraries used in the implementation : Python, TensorFlow-GPU, Keras, Scikit-learn, OpenCV...