[DSRP Evidence](https://dsrpevidence.org/)

# 3D point cloud lithology identification based on stratigraphically constrained continuous clustering

## Details

**Authors** Binqing Gan, Ran Jing, Yanlin Shao, Yuangang Liu, Xiaolei Duan, Peijin Li, Longfan Li

**Year** 2025

**Publisher** Scientific Reports

**Kind of work** article

**Discipline** Geology

**Applied** false

[Read it at the publisher](https://doi.org/10.1038/s41598-025-18946-3) 
10.1038/s41598-025-18946-3

## In authors' words

### Abstract

Three-dimensional laser scanning provides high-precision spatial data for automated lithology identification in geological outcrops. However, existing methods exhibit limited performance in transition zones with blurred boundaries and demonstrate reduced classification accuracy under complex stratigraphic conditions. This study proposes a Stratigraphically Constrained Continuous Clustering (SCCC) framework to address these limitations.

### What they set out to do (purpose)

To develop an automated method for identifying rock types in 3D laser-scanned point clouds of geological outcrops, particularly across blurred stratigraphic transition zones.

### Who or what was studied (sample)

A terrestrial-laser-scanning point cloud of approximately 56 million points from the Qingshuihe Formation outcrop, Junggar Basin, Xinjiang, China.

### How they did it (methods)

A two-stage Stratigraphically Constrained Continuous Clustering framework combining hierarchical clustering constrained by stratigraphic principles with random-forest classification, benchmarked against traditional and deep-learning methods.

### What they found (results)

The stratigraphically constrained clustering method achieved 94.64% overall classification accuracy and 90.87% mean intersection-over-union, exceeding traditional and deep-learning baselines by 26-68%, including in previously error-prone transition zones between rock layers.

## Commentary

### In short

Accurate lithology mapping required explicitly modeling the boundary distinctions between rock units as constrained by their nested position within the larger stratigraphic system.

**Patterns it shows** D, S

**Added** 2026-09-26

**How to cite this** Binqing Gan, Ran Jing, Yanlin Shao, Yuangang Liu, Xiaolei Duan, Peijin Li, Longfan Li (2025). 3D point cloud lithology identification based on stratigraphically constrained continuous clustering. Scientific Reports.