A Review of Software Solutions to Process Ground-based Point Clouds in Forest Applications
Transforming raw point clouds into meaningful forest metrics requires not just advanced sensors, but powerful and adaptable software. As terrestrial (TLS) and mobile laser scanning (MLS) technologies expand in forestry, researchers are collecting more data than ever, billions of points representing trunks, branches, and leaves in extraordinary detail. Yet, data volume alone does not yield insight. Turning these three-dimensional clouds into information usable for forest inventories, ecosystem modeling, and biodiversity studies depends on the tools that process them.
In their 2024 review published in Current Forestry Reports, Murtiyoso et al. provide the first comprehensive assessment of 24 software solutions designed to process ground-based point clouds in forest applications. Conducted under the umbrella of the 3DForEcoTech COST Action, the study maps the fragmented but rapidly evolving landscape of available tools, drawing on contributions from across Europe and beyond. It is both a technical catalogue and a roadmap toward more standardized, interoperable, and user-friendly approaches for precision forestry.
The rise of ground-based remote sensing has created a data explosion. A single TLS setup can capture millions of points per minute, generating gigabytes of data per plot. MLS systems, handheld, backpack, or robot-mounted, extend this capability to dynamic, larger-scale surveys. Together, they have revolutionized how researchers are able to measure trees, estimate biomass, and monitor structure. But this scale of data collection introduces new bottlenecks. Processing point clouds remains complex, often requiring specialized technical expertise and lengthy workflows. Researchers face the challenge of choosing between dozens of tools, many built for specific hardware, formats, or research needs, and integrating them into a coherent pipeline.
Murtiyoso and colleagues highlight a critical need for accessible, standardised, and interoperable workflows. Without shared standards and clear guidance, data processing becomes inefficient and inconsistent, undermining comparability across studies. The review’s primary goal is therefore to identify, compare, and guide end-users from academic researchers to forest managers toward appropriate software for their analytical objectives.
Through a community-driven survey encompassing over 450 participants and expert consultation, the authors identified 24 distinct software solutions for processing ground-based forest point clouds. Of these, 20 are open-source, two are free, and only two are commercial products: a clear indication of the field’s open-science orientation.
The review of the software was divided according to various useful classes. The authors make a distinction between heuristic/rule-based methodology and machine learning approaches. They also provide a useful discussion of detail and scale requirements. Additionally, the paper is a valuable resource for those looking to understand batch processing, file type requirements, and other software intricacies in processing their point clouds.
The general trend of automation at the plot level is identified by the authors. More generally, automation is advancing at every stage, but human expertise remains essential for parameter tuning, validation, and interpretation, particularly in heterogeneous or complex forest environments. While the proliferation of tools demonstrates innovation, it also exposes fragmentation. Murtiyoso et al. emphasize that the field lacks a unified framework for evaluating, comparing, and integrating software.
The authors advocate for open standards and transparent documentation to ensure reproducibility. They also call for benchmark datasets, publicly available, annotated point clouds that can serve as reference material for comparing algorithm performance. Without such common baselines, it remains difficult to assess the reliability or transferability of results across regions, forest types, or sensor platforms. Finally, the review highlights the importance of bridging the gap between forestry science and geomatics software development. Many processing tools originate in photogrammetry, computer vision, or urban mapping communities; adapting them for forestry requires specialized adjustments to account for the irregular geometry of natural vegetation. Stronger cross-disciplinary collaboration could accelerate the translation of advances in 3D data science into practical forestry solutions.
Forest researchers today have an unprecedented toolbox for analysing ground-based point clouds, but it remains scattered. Each software solution contributes a piece of the puzzle, yet without coordination, the overall picture is incomplete. The review by Murtiyoso et al. lays the groundwork for unifying this fragmented ecosystem. By cataloguing available tools, clarifying their roles, and identifying key gaps, it provides a foundation for the next phase: building a standardized, interoperable framework that supports consistent, scalable, and transparent forest data processing.
As laser scanning technologies continue to evolve, so too must the software that interpret them. The path forward lies not just in more sophisticated algorithms, but in shared infrastructure, open standards, benchmark datasets, and collaborative development.
Text is a summarization of a following preprint paper:
Murtiyoso, A., Cabo, C., Singh, A. et al. A Review of Software Solutions to Process Ground-based Point Clouds in Forest Applications. Curr. For. Rep. 10, 401–419 (2024). https://doi.org/10.1007/s40725-024-00228-2
Text is authored by Henry Cerbone – Department of Biology – University of Oxford