DEEP LEARNING AND OPENCV BASED APPROACH FOR REAL TIME OBJECT DETECTION AND TRACKING

Deep learning has had a tremendous impact over the course of the past several years on how the world is reacting to artificial intelligence. Some examples of wellknown object recognition methods (YOLO) include Region-based Convolutional Neural Networks (RCNN), FasterRCNN, Single Shot Detector (SSD), and You Only Look Once. When speed is prioritised over accuracy, YOLO beats the other algorithms, with Faster-RCNN and SSD achieving higher levels of precision than YOLO. Deep learning utilises a combination of SSD and Mobile Netts to increase the effectiveness of detection and tracking operations. This approach effectively detects objects without compromising any speed in the process. In a digital picture or video, an object detection system locates and detects items from the real world that may belong to any class or category, such as people, automobiles, and other vehicles. To complete this objective of finding an item in a photo or video, we used Open-CV packages, convolution neural network (CNN), SVM Classifier, and Evaluation Protocol Map