Abstract

Remote sensing can transform the speed, scale, and cost of biodiversity and forestry surveys. Data acquisition currently outpaces the ability to identify individual organisms in high resolution imagery. We outline an approach for identifying tree-crowns in RGB imagery while using a semi-supervised deep learning detection network. Individual crown delineation has been a long-standing challenge in remote sensing and available algorithms produce mixed results. We show that deep learning models can leverage existing Light Detection and Ranging (LIDAR)-based unsupervised delineation to generate trees that are used for training an initial RGB crown detection model. Despite limitations in the original unsupervised detection approach, this noisy training data may contain information from which the neural network can learn initial tree features. We then refine the initial model using a small number of higher-quality hand-annotated RGB images. We validate our proposed approach while using an open-canopy site in the National Ecological Observation Network. Our results show that a model using 434,551 self-generated trees with the addition of 2848 hand-annotated trees yields accurate predictions in natural landscapes. Using an intersection-over-union threshold of 0.5, the full model had an average tree crown recall of 0.69, with a precision of 0.61 for the visually-annotated data. The model had an average tree detection rate of 0.82 for the field collected stems. The addition of a small number of hand-annotated trees improved the performance over the initial self-supervised model. This semi-supervised deep learning approach demonstrates that remote sensing can overcome a lack of labeled training data by generating noisy data for initial training using unsupervised methods and retraining the resulting models with high quality labeled data.

Highlights

  • Image-based artificial intelligence can advance our understanding of individual organisms, species, and ecosystems by greatly increasing the scale and efficiency of data collection [1]

  • Tree detection is a central task in forestry and ecosystem research and both commercial and scientific applications rely on delineating individual tree crowns from imagery [3,4]

  • Challenges included the over-segmentation of large individual trees, erroneous predicted tree objects based on imperfections in the ground model, and the inclusion of non-tree vertical objects (Figure 5)

Read more

Summary

Introduction

Image-based artificial intelligence can advance our understanding of individual organisms, species, and ecosystems by greatly increasing the scale and efficiency of data collection [1]. Tree detection is a central task in forestry and ecosystem research and both commercial and scientific applications rely on delineating individual tree crowns from imagery [3,4]. While there has been considerable research in unsupervised tree detection while using airborne LIDAR (Light Detection and Ranging; a sensor that uses laser pulses to map three-dimensional structure) [3,5,6], less is known regarding tree detection in RGB (red, green, blue) orthophotos. When combined with hand-crafted rules on tree geometries, these approaches separately performed tree-detection and crown delineation [6,8]. The need to hand-craft tree geometry rules makes it a challenge to create a single approach that encompass a range of tree types [9]

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call