Zheng M.,Beijing University of Technology |
Jia X.,Beijing University of Technology |
Jia X.,Beijing Municipal Key Laboratory
Advances in Intelligent and Soft Computing | Year: 2012
The paper aims to establish a effective feature form of visual speech to realize the Chinese viseme recognition. We propose and discuss a representation model of the visual speech which bases on the local binary pattern (LBP) and the discrete cosine transform (DCT) of mouth images. The joint model combines the advantages of the local and global texture information together, which shows better performance than using the global feature only. By computing LBP and DCT of each mouth frame capturing during the subject speaking, the Hidden Markov Model (HMM) is trained based on the training dataset and is employed to recognize the new visual speech. The experiments show this visual speech feature model exhibits good performance in classifying the difference speaking states. © 2012 Springer-Verlag GmbH.
Kuang J.,Beijing Municipal Key Laboratory |
Shi Y.,Beijing Municipal Key Laboratory |
Cai N.,Beijing Municipal Key Laboratory |
Wang H.,Beijing Municipal Key Laboratory |
Li W.,Beijing Municipal Key Laboratory
Zhongguo Dianji Gongcheng Xuebao/Proceedings of the Chinese Society of Electrical Engineering | Year: 2012
Co-electrolysis of H 2O-CO 2 in solid oxide electrolysis cells is one of the efficient ways to reduce CO 2 emission and to store renewable power. H 2O-CO 2 co-electrolysis performance in a solid oxide electrolysis button cell was tested. Methane composition was discovered in the reduction product gas. The results indicate that the electrochemical performance of H 2O-CO 2 co-electrolysis reaction is between those of steam electrolysis and CO 2 electrolysis, and CO is primarily produced via reverse water gas shift reaction. The concentration of methane in reduction product can be raised by increasing operating voltage or increasing CO 2 concentration in inlet gas. Different with direct H 2-CO methanation, little H 2 can be found in H 2O-CO 2 co-electrolysis product. A reaction pathway of CH 4 production via in-situ H 2O-CO 2 electrochemical conversion on the Ni catalyst surface is proposed. © 2012 Chinese Society for Electrical Engineering.