International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 176 - Number 27 |
Year of Publication: 2020 |
Authors: Shashank Maan, Anamika Tyagi, Shubham Kumar, Abhishek Kumar |
10.5120/ijca2020920280 |
Shashank Maan, Anamika Tyagi, Shubham Kumar, Abhishek Kumar . Security and Surveillance using Computer Vision. International Journal of Computer Applications. 176, 27 ( Jun 2020), 30-35. DOI=10.5120/ijca2020920280
Object detection and Action recognition is one of the key problems in computer vision. Deep learning has gained an enormous influence on how the world is adapting to Artificial Intelligence since the past few years.The project is mainly made for the purpose of security and surveillance. Action recognition is used to monitor the movements of the subject and object detection is used to detect the objects present in the scene.This project is created in the Python programming language with the help of real time computer vision libraries like OpenCv and YOLO. OpenCv is also used for machine learning, and image processing. We have given a dataset of objects which can detect up to 82 objects. OpenCv uses its dnn module to load pre-trained dataset in the code and match it with the data provided by the numpy matrices. Numpy is also a python library used to store the data in arrays. It works by taking the input as a video stream from the webcam/any cam and applying the algorithm on each frame.