Classification of Dielectric Barrier Discharges Using Digital Image Processing Technology

A digital image processing technology-gray level histogram obtained by image processing of the discharge-is proposed to classify the two kinds of dielectric barrier discharge (DBD) modes. With an increase of the applied voltage, frequency, or exposure time, the kurtosis and the skewness of the gray levels decrease, the standard deviation of the gray levels increases significantly in the case of filamentary mode, in contradistinction to the case of homogeneous mode, where its kurtosis, skewness, and standard deviation remain almost constant. The majority of the pixels correspond to near-zero gray levels in the case of the filamentary mode and have a larger gray level for the homogeneous mode. With a decrease of the pressure, the mean gray level in the homogeneous mode increases significantly when the voltage is kept fixed. This suggests that the onset voltage in higher pressure is larger than that in lower pressure. The mean gray level in Ar is larger than that in He when the voltage is kept fixed. It may mean that the onset voltage in He is larger than that in Ar. These results indicate that the method is not only effective but also simple to classify the DBD modes.