Automatic Pith Detection in Tree Cross-Section Images Using Deep Learning

2025-11-29

AI Summary

Manual detection of pith centers in tree cross-sectional images suffers from low efficiency, high subjectivity, and significant measurement errors. To address this, this paper proposes a deep learning–based automated pith detection framework. We systematically evaluate five state-of-the-art models—YOLOv9, U-Net, Swin Transformer, DeepLabV3, and Mask R-CNN—and introduce two key innovations: (1) dynamic data augmentation tailored to anatomical variability across tree species, and (2) a post-processing optimized non-maximum suppression (NMS) strategy for precise localization. Experimental results demonstrate that the Swin Transformer achieves the highest detection accuracy (0.94), while NMS-optimized Mask R-CNN improves mean Intersection-over-Union (mIoU) from 0.45 to 0.80. The method is rigorously validated on multi-source, heterogeneous datasets, exhibiting strong generalization across diverse species and imaging conditions. This work delivers a high-accuracy, robust, and fully automated solution for intelligent forest resource analysis.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Ensemble Methods
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Pith detection in tree cross-sections is essential for forestry and wood quality analysis but remains a manual, error-prone task. This study evaluates deep learning models -- YOLOv9, U-Net, Swin Transformer, DeepLabV3, and Mask R-CNN -- to automate the process efficiently. A dataset of 582 labeled images was dynamically augmented to improve generalization. Swin Transformer achieved the highest accuracy (0.94), excelling in fine segmentation. YOLOv9 performed well for bounding box detection but struggled with boundary precision. U-Net was effective for structured patterns, while DeepLabV3 captured multi-scale features with slight boundary imprecision. Mask R-CNN initially underperformed due to overlapping detections, but applying Non-Maximum Suppression (NMS) improved its IoU from 0.45 to 0.80. Generalizability was next tested using an oak dataset of 11 images from Oregon State University's Tree Ring Lab. Additionally, for exploratory analysis purposes, an additional dataset of 64 labeled tree cross-sections was used to train the worst-performing model to see if this would improve its performance generalizing to the unseen oak dataset. Key challenges included tensor mismatches and boundary inconsistencies, addressed through hyperparameter tuning and augmentation. Our results highlight deep learning's potential for tree cross-section pith detection, with model choice depending on dataset characteristics and application needs.

Problem

• Automating pith detection in tree cross-section images using deep learning models.
• Evaluating model performance on accuracy, segmentation, and boundary precision.
• Addressing challenges like tensor mismatches and improving generalization with augmentation.

Innovation

• Evaluated multiple deep learning models for automated pith detection
• Applied dynamic data augmentation to enhance model generalization
• Used Non-Maximum Suppression to improve Mask R-CNN's detection accuracy