Implementation and optimization of embedded Face Detection system

Aby, P.K; Jose, Anumol; Jose, Bibin; Dinu, L.D; John, Jomon; Sabarinath, G

doi:10.1109/icsccn.2011.6024553

Cited by 10 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main advantage of this method is the fast sum computation using the integral image. It's called Haar-like because it's based on the same principle of Haar wavelets [3].…”

Section: Object Detectionmentioning

confidence: 99%

Multiple object detection using OpenCV on an embedded platform

Guennouni

Ahaitouf

Mansouri

2014

2014 Third IEEE International Colloquium in Information Science and Technology (CIST)

View full text Add to dashboard Cite

Object detection has been attracting much interest due to the wide spectrum of applications that use it. Object detection technology has been driven by an increasing processing power available in software and hardware. In this work we present a developed application for multiple objects detection based on OpenCV libraries. The complexity-related aspects that were considered in the object detection using cascade classifier are described. Furthermore, we discuss the profiling and porting of the application into an embedded platform and comparing the results with the regular platform. The proposed application deals with real time systems implementation and the results give an indication of where the cases of object detection applications may be more complex and where it may be simpler.

show abstract

“…The main advantage of this method is the fast sum computation using the integral image. It's called Haar-like because it's based on the same principle of Haar wavelets [3].…”

Section: Object Detectionmentioning

confidence: 99%

Multiple object detection using OpenCV on an embedded platform

Guennouni

Ahaitouf

Mansouri

2014

2014 Third IEEE International Colloquium in Information Science and Technology (CIST)

View full text Add to dashboard Cite

show abstract

“…1 Standard video input decoding module use video decoding chip TVP5158 as the core, and convert dual PAL analog video signal to a digital signal. It is compatible with the Camera interface of the DM3730 processor [2] and realizes time-sharing input for multiple videos. As the control center and image processing center of the entire system, DM 3730 is responsible for the video signal acquisition and image processing and in the meantime controls the 7-inch LCD module to real-time display the scene and pupil in double window.…”

Section: System Hardware Designmentioning

confidence: 99%

“…In addition, it needs to provide processor 26 MHZ external crystal oscillator, which through the interior phase locked loop frequency doubling for kernel clock. In order to make the system more stable, OMAP processor's ARM kernel and DSP respectively run on rated frequency 1 GHZ and 800 MHZ [2] . chip memory is a single memo ry one which integrated with RAM and FLASH.…”

Section: A Dm3730 Core Modulementioning

confidence: 99%

Design and Implementation of Embedded Pupil Location System

Yan

Ding

2012

2012 Second International Conference on Instrumentation, Measurement, Computer, Communication and Control

View full text Add to dashboard Cite

Based on the needs for the development of the miniaturization and mobility for gaze tracking technology, this paper presents a design proposal about embedded pupil location system. By using processor DM3730 as the core of the hardware platform, we have realized dual-core communication between the ARM and the DS P. This design can make full use of the hardware resources and improve the execution speed of the algorithm. The software graphical interfaces use the methods of interprocess communication. Through calling the executable file generated by ellipse fitting algorithm based on the least-square principle, the accurate location of the pupil can be realized. Experiments show that the system achieved a rate of 25 frames per second for the human eye pupil location, which is a good result and lays the foundation for the implementation of the embedded gaze tracking system.

show abstract

“…Previously studies have investigated the potential of optimising the Viola-Jones algorithm or adjusting it to perform more favourably under differing conditions [1,9,10].…”

Section: Related Workmentioning

confidence: 99%

“…Aby et al [1] explored optimisation of the Viola-Jones algorithm on an embedded, single-board, computing platform -the Beagle Board. Rather than directly optimising the algorithm, they scheduled the heavy computational tasks on the Digital Signal Processor, freeing up the main ARM processor to complete ancillary computational tasks.…”

Section: Related Workmentioning

confidence: 99%

Deep Parameter Optimisation for Face Detection Using the Viola-Jones Algorithm in OpenCV

Bruce¹,

Aitken

Petke³

2016

Search Based Software Engineering

View full text Add to dashboard Cite

Abstract. OpenCV is a commonly used computer vision library containing a wide variety of algorithms for the AI community. This paper uses deep parameter optimisation to investigate improvements to face detection using the Viola-Jones algorithm in OpenCV, allowing a tradeoff between execution time and classification accuracy. Our results show that execution time can be decreased by 48% if a 1.80% classification inaccuracy is permitted (compared to 1.04% classification inaccuracy of the original, unmodified algorithm). Further execution time savings are possible depending on the degree of inaccuracy deemed acceptable by the user.

show abstract

Implementation and optimization of embedded Face Detection system

Cited by 10 publications

References 4 publications

Multiple object detection using OpenCV on an embedded platform

Multiple object detection using OpenCV on an embedded platform

Design and Implementation of Embedded Pupil Location System

Deep Parameter Optimisation for Face Detection Using the Viola-Jones Algorithm in OpenCV

Contact Info

Product

Resources

About