High speed genetic lips detection by dynamic search domain control (特集:システム制御のための画像応用技術)
スポンサーリンク
概要
- 論文の詳細を見る
In this paper, high-speed size and orientation invariant lips detection of a talking person in an active scene using template matching and genetic algorithms is proposed. As part of the objectives, we also try to acquire numerical parameters to represent the lips. The information is very important for many applications, where high performance is required, such as audio-visual speech recognition, speaker identification systems, robot perception and personal mobile devices interfaces. The difficulty in lips detection is mainly due to deformations and geometric changes of the lips during speech and the active scene by free camera motion. In order to enhance the performance in speed and accuracy, initially, the performance is improved on a single still image, that is, the base of video processing. Our proposed system is based on template matching using genetic algorithms (GA). Only one template is prepared per experiment. The template is the closed mouth of a subject, because the application is for personal devices. In our previous study, the main problem was trade-off between search accuracy and search speed. To overcome this problem, we use two methods: scaling window and dynamic search domain control (SD-Control). We therefore focus on the population size of the GA, because it has a direct effect on search accuracy and speed. The effectiveness of the proposed system is demonstrated by performing computer simulations. We achieved a lips detection accuracy of 91.33% at an average processing time of 33.70 milliseconds per frame.
- 社団法人 電気学会の論文
著者
-
Tanaka Kanya
Graduate School Of Sci. And Engineering Yamaguchi Univ.
-
Tanaka Kanya
Graduate School Of Science And Engineering Yamaguchi University
-
Wakasa Yuji
Graduate School Of Sci. And Engineering Yamaguchi Univ.
-
AKASHI Takuya
Graduate School of Science and Engineering, University of Yamaguchi
-
Karungaru Stephen
Institute of Technology and Science, The University of Tokushima
-
Fukumi Minoru
Institute of Technology and Science, The University of Tokushima
関連論文
- PID controller tuning based on the covariance matrix adaptation evolution strategy (特集 進化技術の基礎と応用)
- Hour-glass neural network based daily money flow estimation for automatic teller machines (特集 平成20年電気学会電子・情報・システム部門大会)
- Interactive interface with evolutionary eye sensing and physiological knowledge (特集 平成20年電気学会電子・情報・システム部門大会)
- FRIT for Systems with Dead-Zone and Its Application to Ultrasonic Motors
- High speed genetic lips detection by dynamic search domain control (特集:システム制御のための画像応用技術)
- Genetic Eye Detection Using Artificial Template (Special Issue on Nonlinear Circuits and Signal Processing)
- A CONTROLLER PARAMETER TUNING METHOD CONSIDERING DEAD-ZONE AND SATURATION AND ITS APPLICATION TO ULTRASONIC MOTORS
- FRIT for Systems with Dead-Zone and Its Application to Ultrasonic Motors
- Stochastic Asymptotic Stabilizers for Deterministic Input-Affine Systems Based on Stochastic Control Lyapunov Functions
- Fixed-Structure H_∞ Controller Synthesis Based on the Covariance Matrix Adaptation Evolution Strategy
- A Distributed Consensus Algorithm via LMI-Based Model Predictive Control and Primal-Dual Decomposition
- Controller Parameter Tuning for Systems with Hysteresis and Its Application to Shape Memory Alloy Actuators