A SURVEY ON OBJECT DETECTION TECHNIQUES USING TENSORFLOW, KERAS AND YOLO

Computer Engineering

This paper aims to discuss about some project works that implements CNN with the help of TensorFlow, Keras and YOLO object detection and improvements in object detection. TensorFlow with Keras is a very famous machine learning system and used by various researchers in data classification areas. In TensorFlow computation is represented in the form of dataflow graphs. With the help of TensorFlow, Google’s library a complex computation can be performed very easily by representing them in the form of graph and efficiently mapping the graph parts to machine in the form of cluster. You Only Live Once (YOLO) is an effective real time object detection system. It considers object detection as a regression problem and finds the class probabilities for each of the bounding boxes. Technology has advanced tremendously over the past century, everything starting from the Internet of Things (IoT) to machine learning and deep learning. CNN is used in various fields like medical, marine science and many other applications and has become a prominent domain of machine learning. Implementation of TensorFlow, Keras and YOLO object detection methods gives more accuracy, robustness and faster detection

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.