WlCount: Geological lamination detection and counting using an image analysis approach. (March 2022)

• Record Type: Journal Article
• Authors: Oriani, Fabio; Treble, Pauline C.; Baker, Andy; Mariethoz, Gregoire
• Is Part Of: Computers & geosciences. Volume 160(2022)
• Publication Date: 2022-03
• Subjects: Laminae count -- Stalagmite -- Tree rings -- Geochronology -- Dynamic time warping; Environmental policy -- Periodicals; 550.5
• DOI: 10.1016/j.cageo.2022.105037
• Languages: English
• ISSNs: 0098-3004

Abstract

Abstract: The manual identification and count of laminae in layered textures is a common practice in the study of geological records, which can be time consuming and carry large uncertainty for dense or disturbed lamina textures. We present here a novel image analysis approach to detect and count laminae in geoscientific imagery, called WlCount. Based on Dynamic Time Warping and Wavelet analysis, WlCount firstly aligns persistent vertical elements to increase the continuity of the lamina structure. Then, using a graphical interface, the user extracts the most significant signal frequencies and allows the automatic count of the laminae. The software, tested on a series of stalagmite cut images showing different types of laminations and a tree-ring image, provides an estimation of the laminae detection and count comparable to the manual one. WlCount presents as a useful open-source tool to help geoscientists, sensibly speeding up the lamination count process. Highlights: A new technique for laminae or tree-ring detection and counting. Dynamic Time Warping used to extract laminae time series from entire 2D images. A GUI allows the user operating a semi-automatic count with no parameter setup. The technique is adaptive to complex and weakly defined laminated structures