Ground filtering is crucial for airborne Light Detection and Ranging (LiDAR) data post-processing. The progressive triangulated irregular network densification (PTD) algorithm and its variants outperform others in accuracy, stability, and robustness, using grid-based seed point selection, TIN construction, and iterative rules for ground point identification. However, these methods still face limitations in removing low points and accurately preserving terrain details, primarily due to their sensitivity to grid size. To overcome this issue, a novel PTD filtering algorithm based on an adaptive grid (AGPTD) was proposed. The main contributions of the proposed method include an outlier removal method using a radius outlier removal algorithm and Kd-tree, a method for establishing an adaptive two-level grid based on point cloud density and terrain slope, and an adaptive selection method for angle and distance thresholds in the iterative densification processing. The performance of the AGPTD algorithm was assessed based on widely used benchmark datasets. Results show that the AGPTD algorithm outperforms the classical PTD algorithm in retaining ground feature points, especially in reducing Type I error and average total error significantly. In comparison with other advanced algorithms developed in recent years, the novel algorithm showed the lowest average Type I error, the minimal average total error, and the greatest average Kappa coefficient, which were 1.11%, 2.28%, and 90.86%, respectively. Additionally, the average accuracy, precision, and recall of AGPTD were 97.69%, 97.52%, and 98.98%, respectively.

Point cloud filtering is a vital post-processing step of airborne Light Detection and Ranging (LiDAR) data for multiple applications. Many point cloud filtering algorithms have been proposed. Among them, the progressive triangulated irregular network densification (PTD) algorithm and its improved versions possess certain advantages in terms of algorithm accuracy, stability, and robustness. However, these methods still face limitations in accurately preserving terrain details and effectively eliminating low points in regions characterized by discontinuous terrain and steep slopes, largely due to their sensitivity to the size of the grid. To overcome this issue, a novel PTD filtering algorithm based on an adaptive grid (AGPTD) was proposed. The main idea is to extract more initial ground seed points at locations with discontinuous terrain by adaptively establishing a secondary grid based on point cloud density and terrain slope, to better preserve the terrain features for the subsequent iterative densification processing. The performance of the AGPTD algorithm is assessed based on widely used benchmark datasets. Results show that the AGPTD algorithm outperforms the classical PTD algorithm in retaining the ground feature points, especially in reducing the Type I error and average total error significantly. In comparison with other advanced algorithms developed in recent years, the novel algorithm shows the lowest average Type I error, the minimal average total error, and the greatest average Kappa coefficient, which are 1.22%, 2.31%, and 90.94%, respectively.

Sunspots and filaments are critical indicators of solar activity. While the Chinese H α Solar Explorer (CHASE) provides massive full-disk spectral data, existing automated detection algorithms are primarily optimized for other solar observatories, such as the Solar Dynamics Observatory/Helioseismic and Magnetic Imager. The significant discrepancies in instrument responses between different solar observatories render the direct transfer of existing algorithms ineffective, failing to yield robust results when processing CHASE data. To address these challenges, an automatic detection framework based on an Improved Simulated Annealing Genetic Algorithm (ISAGA) is proposed. First, a standardized preprocessing system tailored to CHASE imaging characteristics is established to effectively eliminate limb darkening and transmission artifacts in the Fe I and H α bands. Second, the ISAGA algorithm introduces a physical prior reward mechanism based on the Sigmoid function into the conventional Kapur–Sahoo–Wong (KSW) entropy model, to overcome the shortcomings of handling unimodal histograms, significantly improving the threshold search robustness. Finally, topological postprocessing is used to achieve refined feature extraction. Experiments on the 2022–2023 CHASE dataset demonstrate that ISAGA achieves an intersection over union (IoU) of 98.43% and an F1 score of 99.20% for sunspot detection, outperforming other representative methods. For solar filaments, the framework yields an average IoU of 60.93% and an F1 score of 74.95%. By achieving an effective balance between superior precision and high computational efficiency, ISAGA is suitable for the automated detection of sunspots in massive CHASE data and serves as an effective tool for the rapid generation of large-scale filament datasets.