Face Detection Based on Viola-Jones Algorithm Applying Composite Features

Viola-Jones' face detection algorithm was jointly proposed by Paul Viola and Michael Jones. Although it realized face real-time detection to some extent, its false detection rate is not low. Because the block features in the Viola-Jones algorithm can't handle purely rigid objects, such as...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2019 International Conference on Robots & Intelligent System (ICRIS) s. 82 - 85
Hlavní autoři: LU, Wen-yao, YANG, Ming
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.06.2019
Témata:
Classification algorithms
Compound Features
Compounds
Face
Face detection
Face recognition
Feature extraction
Linear discriminant analysis
Viola-Jones
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Viola-Jones' face detection algorithm was jointly proposed by Paul Viola and Michael Jones. Although it realized face real-time detection to some extent, its false detection rate is not low. Because the block features in the Viola-Jones algorithm can't handle purely rigid objects, such as chopsticks and cups, so if there are rigid objects in the face image, Viola-Jones' face detection algorithm is prone to generate false detection of faces. In this paper, we propose to apply the composite features based on Viola-Jones algorithm to improve the above problems, and prove the feasibility of this method through experiments.
DOI:10.1109/ICRIS.2019.00029