Open AccessObject Detection using OpenCV and Deep Learning
Author(s) -
G. Balaji
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.35880
Subject(s) - computer science , artificial intelligence , object detection , minimum bounding box , categorization , computer vision , object (grammar) , pattern recognition (psychology) , class (philosophy) , bounding overwatch , population , detector , single shot , volume (thermodynamics) , image (mathematics) , telecommunications , physics , demography , quantum mechanics , sociology , optics
Object Detection using SSD (Single Shot Detector) and MobileNets are efficient because this technique detects objects quickly with less resourses without sacrificing performance. In this every class of item for which the classification algorithm has been trained generates a bounding box and an annotation describing that class of object. This provides the foundation for creating several types of analytical features such as the volume of traffic in a certain area over time or the entire population in an area is real-time detection and categorization of objects from video data.