Image

Revolutionising forestry with 3D deep learning

3 October 2023

In the rapidly evolving world of artificial intelligence (AI) and deep learning, one of the most critical factors for success is access to vast amounts of labelled data. These labelled datasets power AI models by giving them lots of examples to learn from. While researchers and data scientists in many industries have enjoyed the benefits of extensive labelled datasets to train and finetune their deep learning models, forestry has often lagged behind due to the absence of large, labelled datasets.

Now, thanks to an international collaboration involving Scion, the University of Gottingen and the University of Applied Science and Arts (HAWK) in Germany and led by Dr Stefano Puliti at the Norwegian Institute of Bioeconomy Research (NIBIO), a benchmark dataset for forestry is publicly accessible.

Research papers associated with the two-year project have recently been published. However, the impact of the ‘For-Instance’ dataset extends far beyond the academic realm. More than simply a collection of labelled data, the dataset marks a significant turning point in 3D deep learning for forestry by providing researchers and AI specialists with the raw material they need to train and test their models on real-world forest data for the first time.

At a time when data is often considered the most valuable commodity, this dataset breaks down the barriers to entry for AI adoption in forestry. Instead of spending months collecting and labelling data themselves, data scientists at New Zealand companies and industry service providers now have access to a valuable resource that can accelerate their own innovative AI-driven solutions for forestry. Scion’s team lead for Remote Sensing and GIS, Grant Pearse, says the dataset is “game-changing” for New Zealand’s forestry industry as it will unlock a multitude of possibilities.

“We contributed labelled data from radiata pine to empower researchers and industry to develop and test models specifically for point clouds collected from our commercial radiata forests. The data are labelled in such a way that models can be trained to identify and segment individual trees as well as the stems, live branches and woody branches of each tree within the lidar point clouds – this directly aligns the AI models with the needs of New Zealand’s forestry sector.

“But what makes this dataset even more powerful is its benchmarking aspect.” In machine learning and AI, benchmark datasets play a pivotal role in advancing research. They provide a standardised way to evaluate different algorithms and approaches. The FOR-Instance dataset not only offers labelled data, but also predefines how the data should be split for training and evaluation of AI models. This ensures that researchers and industry can make fair and objective comparisons between their models and others in the field, enhancing the credibility of their work.

More >>

As part of the upcoming ForestTECH 2023 event, Grant Pearse, Scion’s Team Leader for Remote Sensing & GIS will be outlining to local foresters as part of the NZ leg of the annual series on 14-15 November, opportunities for using one of the largest datasets for high resolution land-cover mapping (a digital twin of New Zealand’s productive forest estate at national scale). Full details on the ForestTECH 2023 programme can be found here.

Source: Scion, FIEA

Read More
Image

New tool to manage forestry slash

Better management of forest felling waste has gained new impetus in Cyclone Gabrielle’s wake, with the sector firmly in the spotlight for the impact such waste has had on communities, ecosystems, and farmland downstream from harvest areas. Rotorua-based forest management ...
Read more
Image

New multi-species carbon calculator available

A new multi-species carbon calculator developed by Mark Kimberley and Michael Watt to support grower decision-making is freely available on the Forest Growers Research website ( Multi-Species Carbon Calculator (fgr.nz)). The calculator predicts growth metrics for eleven species that include ...
Read more
Image

JOUAV unveils advanced airborne Lidar sensor

JOUAV has released the JoLiDAR-1000 Lidar sensor designed specifically for uncrewed aerial vehicles (UAVs or ‘drones’) as the latest addition to its array of high-performance Lidar sensors for use in UAV applications such as GIS, surveying and precision powerline inspections. ...
Read more
Image

Wingtra debuts pioneering Lidar drone solution

Wingtra has unveiled its innovative Lidar-drone solution. The new Lidar payload is designed to seamlessly integrate the efficiency, ease of operation, and high accuracy of the WingtraOne GEN II drones with an advanced Wingtra-engineered Lidar sensor, offering notable advancements in ...
Read more
Image

Trimble and DroneDeploy improved UAV mapping

Trimble has unveiled the integration of the Trimble Applanix POSPac Cloud post-processed kinematic (PPK) GNSS positioning service, featuring CenterPoint RTX, with the UAV-based mapping and data collection capabilities of DroneDeploy’s reality capture platform. With the Trimble cloud positioning service, DroneDeploy ...
Read more
Image

Blue gum expansion & mechanised planting planned

Australia’s dwindling timber reserves could be exhausted in less than a decade, but industry experts are hoping a new native nursery in Western Australia could help ease the pressure. In the coastal city of Albany, 450 kilometres south of Perth, ...
Read more