Spatial frequency based video stream analysis for object classification and recognition in clouds

Abstract

The recent rise in multimedia technology has made it easier to perform a number of tasks. One of these tasks is monitoring where cheap cameras are producing large amount of video data. This video data is then processed for object classification to extract useful information. However, the video data obtained by these cheap cameras is often of low quality and results in blur video content. Moreover, various illumination effects caused by lightning conditions also degrade the video quality. These effects present severe challenges for object classification. We present a cloud-based blur and illumination invariant approach for object classification from images and video data. The bi-dimensional empirical mode decomposition (BEMD) has been adopted to decompose a video frame into intrinsic mode functions (IMFs). These IMFs further undergo to first order Reisz transform to generate monogenic video frames. The analysis of each IMF has been carried out by observing its local properties (amplitude, phase and orientation) generated from each monogenic video frame. We propose a stack based hierarchy of local pattern features generated from the amplitudes of each IMF which results in blur and illumination invariant object classification. The extensive experimentation on video streams as well as publically available image datasets reveals that our system achieves high accuracy from 0.97 to 0.91 for increasing Gaussian blur ranging from 0.5 to 5 and outperforms state of the art techniques under uncontrolled conditions. The system also proved to be scalable with high throughput when tested on a number of video streams using cloud infrastructure.

Deep Learning Hyper-Parameter Optimization for Video Analytics in Clouds

“…Recent video analytics systems often use shallow networks and hand crafted features to perform object classification [30] [31]. These hand crafted features are combined to generate larger features.…”

Edge Enhanced Deep Learning System for Large-Scale Video Stream Analytics

“…Recently intelligent video analytics systems (IVS) with the aim to analyze video streams without human intervention have emerged which provide constant analysis of a scene [10]. Most of the existing work on intelligent video analysis is based on centralized architecture using a storage-analyze cloud model such as [11] where the data is first transported and stored in the cloud. The analytics is then performed on the stored data using job scheduling [12].…”

“…where A t is the algorithm time to process the job and T t is the network time to transfer the record from source to destination. Total cost to process all the jobs with algorithm A will be C T ja = J T jps * C ja (11) where J T jps is the total number of jobs and C ja is the cost time for each job. We also define a percentage gain to show the efficiency of a setup S in terms of time and cost of a reference cloud setup.…”

Cloud‐based video analytics using convolutional neural networks

“…A number of studies worked on multimodal features. They used hidden Markov models, Guassian density functions, and local pattern features in their system.…”