Deep Learning Models for Analysis of Traffic and Crowd Management from Surveillance Videos
| Autor: | S. Seema, Suhas Goutham, Smaranita Vasudev, Rakshith R. Putane |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Artificial neural network
Computer science business.industry Deep learning Real-time computing ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Problem statement Optical character recognition Object (computer science) computer.software_genre Object detection Field (computer science) Tesseract Artificial intelligence business computer |
| Zdroj: | Advances in Intelligent Systems and Computing ISBN: 9789811524134 |
| DOI: | 10.1007/978-981-15-2414-1_9 |
| Popis: | Deep learning models have been used in the field of object detection and object counting. The problem statement dealt with in this paper aims to achieve the objectives of traffic and crowd management. The Single Shot MultiBox Detector (SSD) model is used in conjunction with a line of counting approach to count the objects of interest in a video captured using surveillance cameras. The proposed model has been used for analyzing traffic surveillance videos to make intelligent traffic decisions to prioritize traffic signals based on the traffic densities. As a sub case of traffic management, a Tesseract OCR model is used to capture the license plate of vehicles violating any traffic regulations. For crowd management, surveillance videos are analyzed to obtain the crowd statistics to handle crowd management in cases of emergencies and huge public gatherings for safety and security. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|