International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume VII, Issue VIII, August 2022 | ISSN 2454–6194
Identifying Objects in Real-Time at the Lowest Framerate
Md. Mamun Hossain*, Md. Ashiqur Rahman, Humayra Ahmed
Department of Computer Science and Engineering,
Bangladesh University of Business and Technology, Dhaka, Bangladesh
*Corresponding Author
Abstract: The practice of finding instances of semantic objects of a certain class, including people, cars, and traffic signs, in digital photos and videos is known as object identification or detection. Due to the development of high-resolution cameras and their widespread usage in everyday life, the detection is one of the most difficult and rapidly expanding study fields in computer science, particularly in computer vision. For automatic object recognition, several researchers have experimented with a variety of techniques, including image processing and computer vision. In this research, we employed a deep learning based framework YOLOv3 using Python, Tensorflow, and OpenCV to identify objects in real time. We do a number of tests using the COCO dataset to verify the effectiveness of the suggested strategy. The results of the experiments show that our suggested solution is resource and cost effective since it uses the fewest frames per second.
Index Terms: Object recognition, Realtime, YOLOv3, Tensor- flow, COCO dataset
I. INTRODUCTION
Over the past several years, object detection has had a significant impact on how the world has adapted to artifi- cial intelligence. Real-time object detection is essential in autonomous systems that are Computer Vision (CV) capable. Its precision and speed are equally crucial for ensuring reliable functioning. Although object recognition for static images has been extensively investigated, real-time object detection presents a number of distinct difficulties, including motion blur produced on by moving objects, focusing issues, and real- time speed limitations for autonomous agents. Real-time object identification, however, also creates fresh opportunities that may be taken advantage of. The key findings is that when it comes to image scaling, accuracy and speed should not always be traded off. Our findings demonstrate that occasionally real- time object recognition accuracy is improved by downscaling the image to a lower resolution. Future autonomous agents like self-driving vehicles, drones, and robots need real-time object detection as a crucial building component for visual cognition. Therefore, it is crucial for the