Downhole image processing method based on improved bilateral filtering

Abstract

The invention discloses a downhole image processing method based on improved bilateral filtering. According to the method, improved bilateral filtering is adopted to estimate an illumination image, an Otsu threshold segmentation method is adopted to obtain a reflection image threshold, bilateral filtering is adopted to perform denoising processing of different degrees on bright and dark areas, segmented linear transformation is adopted to enhance and suppress the bright and dark areas, and the image contrast is improved. Experimental results show that the brightness and contrast of the coal mine underground low-illumination image can be effectively enhanced, and noise can be effectively removed while underground dark region details are ensured. The problems of halo and ash flooding and noise amplification in the traditional method are avoided.

Description

Technical field:

[0001] The invention belongs to the field of image processing, in particular to an underground image processing method based on improved bilateral filtering. Background technique:

[0002] Due to the particularity of the downhole working environment, in the process of monitoring downhole production, the quality of the monitoring image is often affected by the harsh downhole environment, with low contrast, low illumination, blurred details in dark areas and a lot of noise. , so it is necessary to perform some processing on the downhole image to enhance the visual effect of the image.

[0003] The traditional methods of image enhancement include histogram equalization method, dark channel prior method, unsharp mask method, etc. The histogram equalization method is simple and efficient, and the processing speed is fast, but at the same time, the noise of the image is amplified in the process of processing, and the image details are blurred and the contrast is ...

Images