TORNADO: Intermediate Results Orchestration based Service-Oriented Data Curation Framework for Intelligent Video Big Data Analytics in the Cloud.

Aftab Alam,Young-Koo Lee

doi:10.3390/s20123581

Abstract

In the recent past, the number of surveillance cameras placed in the public has increased significantly, and an enormous amount of visual data is produced at an alarming rate. Resultantly, there is a demand for a distributed system for video analytics. However, a majority of existing research on video analytics focuses on improving video content management and rely on a traditional client/server framework. In this paper, we develop a scalable and flexible framework called TORNADO on top of general-purpose big data technologies for intelligent video big data analytics in the cloud. The proposed framework acquires video streams from device-independent data-sources utilizing distributed streams and file management systems. High-level abstractions are provided to allow the researcher to develop and deploy video analytics algorithms and services in the cloud under the as-a-service paradigm. Furthermore, a unified IR Middleware has been proposed to orchestrate the intermediate results being generated during video big data analytics in the cloud. We report results demonstrating the performance of the proposed framework and the viability of its usage in terms of better scalability, less fault-tolerance, and better performance.

Highlights

Videos are recorded and uploaded to the cloud regularly
A unified Intermediate Results (IR) Middleware has been proposed to orchestrate the intermediate results being generated during video big data analytics in the cloud
The existing solutions do not consider factors like the management of high-level and low-level features while deploying Intelligent Video Analytics (IVA) algorithms. Motivated by these limitations in existing work, we propose and implement a comprehensive intermediate results orchestration based service-oriented data curation framework for large-scale online and offline IVA in the cloud known as TORNADO

Summary

Introduction

Sources that are actively contributing to video generation include CCTV, smartphones, drones, and many more, which have resulted in a big data revolution in video management systems. Devices will see a growth rate of 28.7% over the period 2018–2025, where surveillance videos are the majority shareholder, i.e., 65% [3]. Such an enormous video data is considered as “video big data” because a variety of sources generate a large volume of video data at high velocity that holds high value. Video big data pose challenges for video management, processing, mining, and manipulation. Layer (VDPL) is in charge of pre-processing and extracting the significant features from the raw videos and input to the Video Data Mining Layer (VDML).

Objectives

Results

Conclusion