EGS-CSF: an adaptive ground point cloud filtering framework based on evolutionary gradient search
To address the limitations of traditional Cloth Simulation Filtering (CSF) algorithms – namely, high parameter sensitivity and dependence on manual tuning in complex terrains – we propose EGS-CSF, an adaptive framework based on Evolutionary Gradient Search (EGS). By integrating global exploration wi...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/17538947.2025.2531843 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | To address the limitations of traditional Cloth Simulation Filtering (CSF) algorithms – namely, high parameter sensitivity and dependence on manual tuning in complex terrains – we propose EGS-CSF, an adaptive framework based on Evolutionary Gradient Search (EGS). By integrating global exploration with local convergence, EGS-CSF dynamically optimizes key parameters (R, S, D, T), substantially reducing filtering errors. Experiments on the OpenGF and ISPRS datasets, which include steep slopes, urban areas, and vegetated regions, demonstrate that EGS-CSF achieves an average total error of 2.65% across 15 samples and over 90% accuracy on the OpenGF dataset, outperforming standard CSF by 10–20%. These results highlight the robustness and effectiveness of the proposed method across diverse terrain conditions. |
|---|---|
| ISSN: | 1753-8947 1753-8955 |