Image analysis techniques and gray-level co-occurrence matrices (GLCM) for calculating bioturbation indices and characterizing biogenic sedimentary structures

Aspects of texture and structure in a bed resulting from bioturbation can provide valuable information about the ecology and environment at the time of deposition. However, not only the degree of bioturbation, but the structure of the burrows is important for interpreting biogenic fabrics. Here, image analysis is applied to real and artificial images of biogenic sedimentary structures. Image segmentation was applied to images of Middle Ordovician biogenic sedimentary structures from Dixon, Illinois (Pecatonica Formation), isolating the biogenic sedimentary structures. A gray-level co-occurrence matrix (GLCM) is calculated from the segmented image and eight artificial images representing different levels of image noise. Texture measures were calculated from the GLCMs and compared with identify scale and directional structural differences between the images. Principal component analysis was used to statistically group the images. Artificial images were found to be distinguishable from the real images by GLCM texture measures, and the real images differed most significantly at the largest scales.