A Combined Noisy Borehole Image Log Segmentation Method
Letı́cia da Silva Bomfim, Hélio Pedrini, Alexandre Campane Vidal · 2023
Borehole Image Logs are a valuable tool for studying and characterizing reservoirs. However, the quality of these images is often compromised by various types of noise due to the environment in which they are acquired. Additionally, extracting the relevant information from these images can be a challenging task, as the existing noise can impede data segmentation even after improvement. Therefore, this paper aims to investigate the best combined strategy for enhancing these images using Equalization and Retinex methods, as well as segmenting them using threshold and quantization techniques to achieve better extraction of regions of interest. By analyzing the source of the noise, we determined that the Multi-Scale Retinex with Color Preservation and Color Image Quantization is an effective approach for this task.