Fast Facial Detection by Depth Map Analysis

Ming-Yuan Shieh and Tsung-Min Hsieh

Mathematical Problems in Engineering, 2013, vol. 2013, 1-10

Abstract:

In order to obtain correct facial recognition results, one needs to adopt appropriate facial detection techniques. Moreover, the effects of facial detection are usually affected by the environmental conditions such as background, illumination, and complexity of objectives. In this paper, the proposed facial detection scheme, which is based on depth map analysis, aims to improve the effectiveness of facial detection and recognition under different environmental illumination conditions. The proposed procedures consist of scene depth determination, outline analysis, Haar-like classification, and related image processing operations. Since infrared light sources can be used to increase dark visibility, the active infrared visual images captured by a structured light sensory device such as Kinect will be less influenced by environmental lights. It benefits the accuracy of the facial detection. Therefore, the proposed system will detect the objective human and face firstly and obtain the relative position by structured light analysis. Next, the face can be determined by image processing operations. From the experimental results, it demonstrates that the proposed scheme not only improves facial detection under varying light conditions but also benefits facial recognition.

Date: 2013
References: Add references at CitEc
Citations:

Downloads: (external link)
http://downloads.hindawi.com/journals/MPE/2013/694321.pdf (application/pdf)
http://downloads.hindawi.com/journals/MPE/2013/694321.xml (text/xml)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hin:jnlmpe:694321

DOI: 10.1155/2013/694321

Access Statistics for this article

More articles in Mathematical Problems in Engineering from Hindawi
Bibliographic data for series maintained by Mohamed Abdelhakeem ().