摘要
介质阻挡放电能够高效产生低温等离子体,其放电模式及程度影响着实际工业应用效率。提出一种基于可见光图像的介质阻挡放电(DBD)空间均匀性评价方法,通过计算不同放电条件以及曝光时间下的放电图像的灰度共生矩阵(GLCM)及其4种特征参数,对DBD放电的空间均匀性进行评价;建立不同均匀程度丝状放电的分布模型,根据放电图像GLCM的特征参数判别3种程度的放电等级:Ⅰ级放电分布均匀、Ⅱ级放电分布欠均匀、Ⅲ级极不均匀。得出如下结论:在拍摄曝光时间及实验条件相同时,放电图像的灰度共生矩阵的特征参数可以用来评价介质阻挡放电的均匀程度,这些参量能够很好地反映放电图像在长时间尺度上的特征性质,并且可用于分类器的输入量以作进一步分类;拍摄放电图像曝光时间不同,特征参数有较大变化,在0.3~1 s能更有效地反映放电在空间上的分布情况;在丝状放电状态下,随着电压等级的提升,放电次数会增多,在长时间尺度上空间均匀性更好。
The efficient generation of low-temperature plasma through dielectric barrier discharge(DBD)is influenced by the discharge mode and level,which in turn affects the efficiency of practical industrial applications.A new method is used in to evaluate spatial uniformity of dielectric barrier discharge based(DBD)on visible light images.The spatial uniformity of DBD is evaluated by calculating the gray-level co-occurrence matrix(GLCM)of the discharge image under different discharge conditions,exposure time,and characteristic parameters.Ac⁃cording to the establishment of distribution model of filamentary discharge with different uniform degrees,discharge levels can be determined by the characteristic parameters of GLCM of discharged images:evenly distributed discharge assigned to grade I,unevenly distributed dis⁃charge defined as grade II,and extremely uneven discharge classified as grade III.The results showed that the characteristic parameters of GL⁃CM of discharge image is valid to evaluate the uniformity of DBD under the same exposure time and experimental conditions.These parameters well reflect the characteristics of the discharge image on a long-time scale,and can be used for further classification of the input of the classifi⁃er.The characteristic parameters vary greatly with the exposure time of the captured discharge image,which effectively reflect the spatial dis⁃tribution of discharge in 0.3-1s.With the increase of voltage level,the discharge times of filamentary discharge increased,and the spatial uni⁃formity was improved in the long-time scale.
作者
付袈铭
FU Jiaming(School of Mechanical Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《软件导刊》
2023年第7期144-150,共7页
Software Guide
关键词
介质阻挡放电
均匀性评价
丝状放电
灰度共生矩阵
图像分类
dielectric barrier discharge
uniformity evaluation
filamentous discharge
gray level co-occurrence matrix
image classifica⁃tion
