A Spatial Fusion Enhancement Algorithm for Underground Images

Aiming at the problems of uneven image illumination, strong noise, and blurred edges caused by the acquisition environment or point light sources of downhole images, a spatial fusion enhancement algorithm for downhole images is proposed. First, HSV transforms the source image to obtain the luminance component, and uses a smoothing guided filter to denoise it, perform edge feature extraction on the image, and then perform non-downsampling shear wave on the extracted edge feature image, NSST transform and decompose the luminance component after noise reduction. The high-frequency sub-band uses the PCNN model to fuse the maximum value of the total ignition amplitude to fuse the high-frequency features of the image. The low-frequency sub-band is fused using an improved energy maximum method, which is obtained by inverse NSST and HSV inverse transformation. Finally, the image is enhanced. Experiments show that the improved algorithm enriches the local detail information of the downhole image and achieves good results in objective evaluation indicators.