基于中值滤波的汽油机燃烧火焰图像去噪处理

Gasoline Engine Combustion Flame Images Denoising Processing Based on Median Filter

在基于图像的汽油发动机燃烧火焰三维重构技术中,用于重构的二维火焰图像质量直接关系到重构效果的好坏,利用高速摄像机拍摄到的火焰图像含有大量椒盐噪声,因此必须对其进行去噪预处理。传统中值滤波算法在去除噪声的同时,会对图像产生模糊效应,对于具有大量边缘和细节信息的汽油发动机火焰图像不适用。在针对传统中值滤波算法进行研究的基础上,提出了一种改进的中值滤波算法,通过基于图像灰度分布直方图的全局检测法以及基于均值比较的局部检测法来判定待检测像素点是否为噪声点,如果是则对其进行中值滤波,否则保持图像原值。实验结果表明,该算法在有效去除汽油发动机燃烧火焰图像椒盐噪声的同时,能很好地保留图像边缘特征和细节信息。

In the image-based 3D reconstruction technology of the gasoline engine combustion flame, the quality of the two-dimension flame images used in the reconstruction is directly related to the effect of the reconstruction result, so the images captured by the high-speed camera must firstly be denoised. The traditional median filte- ring algorithm is used to get rid of the pepper and salt noise, but bringing blurred effect at the same time. Therefore, it does not apply to gasoline engine flame images with a lot of edges and detail information. An im- proved median filtering algorithm based on the study of traditional median filtering was proposed. Firstly, this al- gorithm determines whether the pixel point is a noise point using the global detection method based on image gray distribution histogram and local detection method based on mean Comparison. If the pixel point is a noise point, then it needs to be denoised by the median filtering, or else it will keep the original value. The experi- mental results show that the algorithm can effectively remove the impulse noise of gasoline engine combustion flame images, and at the same time it can preserve the image edge features and details very well.