Classification and Identification of Oxyacetylene Flame Images Based on Hidden Markov Model
The combustion process of the oxyacetylene flame is a random process. Good classification and identification results are obtained by using the Hidden Markov Model(HMM) which has good simulation ability on the random process. First, the image sequences samples are gotten using CCD camera. Then the characteristic parameters are picked up and sequences of observation vectors are composed after the pretreatment is done on the images. Finally, the HMM is trained for the combustion process. For the image sequences to be identified, their combustion state sequences can be obtained using Viterbi algorithm after the characteristic parameter vectors are picked up, so the combustion process of oxyacetylene can be classified and identified. The experiment proves that, according to the characteristics of the oxyacetylene flame images, using image sequences instead of images as objects and adopting one dimension and multi-step classification method can save time, improve the real-time and the identifying accuracy.
HMM oxyacetylene flame image identification
Zhang Xin Gao Xiuyan Liu Ying
College of Electronic and Information Engineering of Hebei University,Baoding 071002 China College of Electronic and Information Engineering of Hebei University,Baoding 071002 China Hebei Sof
国际会议
厦门
英文
691-695
2010-05-22（万方平台首次上网日期，不代表论文的发表时间）