TreeBio - Preliminary study on traceability of tree logs using digital log end images Masterarbeit

This thesis focuses on the biometric traceability of wood logs. Biometric log traceability is at an early stage and present research treats traceability in the sawmill. State-of-the-art systems use Radio Frequency Identi cation (RFID) or plastic batches to reach log traceability from forest site to further processing companies. This work is a preliminary study on biometric log traceability using digital log end images. Digital images can be taken at little cost and at almost every stage in the log processing chain. Similar to a human ngerprint or iris the annual ring pattern of a cross-section shows a speci c texture pattern containing biometric features. The study is subdivided into a theoretical and a practical part. In the theoretical part basics of log traceability and biometric systems are considered. Furthermore, a comprehensive literature overview on cross-section imaging and biometric systems in other elds of applications forms the basis for the practical part of this study. In the practical part an initial scheme for a biometric log recognition framework (TreeBio) is introduced. By treating pith estimation and cross-section segmentation in rough log end images this work contributes to the development of the TreeBio framework. For both tasks, computational approaches and experiments are presented. The practical part highlights the di culties of biometric log recognition. This makes it possible to present considerations on future work. keywords image analysis, wood logs, log traceability, pith estimation, cross-section segmentation, cross-section analysis