Dust Filtering in LIDAR Point Clouds Using Deep Learning for Mining Applications

Bruno Cavieres; Nicolas Cruz; Javier Ruiz-del-Solar

doi:10.20944/preprints202509.1214.v1

3. Proposed Dust Filtering Method

The proposed dust filtering method is based on the use of a modified version of the PointNet++ architecture [11], which receives point cloud data directly, groups the points hierarchically, computes internal feature representations, and classifies each point as dust or non-dust. For each LIDAR measurement, its 3D position, intensity, and temporal displacement are provided to the network. In the case where the data scan is acquired from a mobile vehicle, odometry information is also considered. Thus, the dust-filtering process takes into account geometric, intensity, and temporal features.

3.1. Neural Architecture: Reduced-PointNet++

The PointNet network, proposed in [10], is able to directly process the spatial coordinates and other attributes of the points to classify and segment a point cloud. Unlike previous approaches that required converting the point clouds into 2D or volumetric representations, PointNet works directly on 3D data without the need for voxelization or projections, which makes it more efficient in terms of memory and processing. PointNet is based on the use of symmetric functions, such as global max-pooling, which allows it to be invariant to the order of the points, a fundamental property in point clouds where there is no predefined order. In addition, the model incorporates learned transformations that align the point cloud before processing, reducing the variability introduced by differences in data orientation. However, due to its reliance on global feature aggregation, PointNet is limited in its ability to capture local relationships between points, which may affect its performance on tasks that require fine-structure information, as in the case of filtering dust measurements.

PointNet++, proposed in [11], improves PointNet by using hierarchical groupings of PointNet points and subnetworks on multiple scales, allowing greater capture of local features and spatial relationships. This modification allowed its use for dust filtering.

In order to achieve real-time processing of large point clouds, a reduced version of PointNet++ was designed in this work. Two big changes were made to PointNet++: first, the number of sampled points was reduced, which also reduced the number of abstraction layers (SA), and some blocks were eliminated to make the architecture simpler. Second, the MLPs configurations in both the abstraction and propagation blocks were adjusted. This led to a significant decrease in the number of parameters and an improvement in the efficiency of the model. Table 1 shows these changes.

In Section 4, a comparison of the performance of PointNet++ and its reduced version is presented in dust-filtering tasks.

3.2. Point Cloud Features

Let us consider a point cloud

P^{t}

containing points

p_{i}^{t}

belonging to frame t, where t represents an instant in a temporal sequence of length T. Each point

p_{i}^{t}

is defined by:

p_{i}^{t} = [c_{i}^{t}, f_{i}^{t}],

(3)

where

c_{i}^{t} = (x_{i}^{t}, y_{i}^{t}, z_{i}^{t})

corresponds to spatial data and

f_{i}^{t}

represents a vector of associated attributes, such as intensity (

g_{i}^{t}

) or other properties such as temporal information.

A first alternative to incorporate temporal information is to find the spatial differences between points in consecutive point clouds. For each point

p_{i}^{t}

in

P^{t}

, we look for the nearest point in

P^{t - 1}

. This is done by minimizing the following Euclidean distance:

d_{i j}^{t - 1, t} = ∥ c_{i}^{t} - c_{j}^{t - 1} ∥ = \sqrt{{(x_{i}^{t} - x_{j}^{t - 1})}^{2} + {(y_{i}^{t} - y_{j}^{t - 1})}^{2} + {(z_{i}^{t} - z_{j}^{t - 1})}^{2}} .

(4)

Then, the nearest point

p_{j^{*}}^{t - 1}

is determined as:

j^{*} = arg min_{j} d_{i j}^{t - 1, t},

(5)

and the spatial difference vector is computed as:

Δ c_{i}^{t} = c_{i}^{t} - c_{j^{*}}^{t - 1} .

(6)

The vector of spatial differences

Δ c_{i}^{t}

, or its magnitude

m_{i}^{t}

, can then be used as a temporal attribute of each point.

A second alternative to incorporate temporal information, proposed in [12] as temporal variation-aware interpolation, is to generate an interpolated feature to represent the local information of the previous point cloud

P^{t - 1}

projected in the current point cloud

P^{t}

. To do this, first, for each point

p_{i}^{t}

in

P^{t}

, we calculate the distances

d_{i j}^{t - 1, t}

to the K nearest neighbors in

P^{t - 1}

. Then, the interpolation weights for each neighbor point are computed as follows:

w_{i j}^{t - 1, t} = β \cdot (α - min (d_{i j}^{t - 1, t}, α)) .

(7)

with

d_{i j}^{t - 1, t}

defined by (4) and

α

and

β

hyperparameters.

The weights are then normalized using the softmax function:

{\hat{w}}_{i j}^{t - 1, t} = \frac{exp (w_{i j}^{t - 1, t})}{\sum_{k = 1}^{K} exp (w_{i k}^{t - 1, t})} .

(8)

Then, for each of the nearest neighbor points of

p_{i}^{t}

in the previous point cloud, the intensity value and the differences of the intensity values are fed to an MLP layer with ReLU activation, and intermediate features are computed (see details of the network architecture in [12]):

v_{i j}^{t - 1, t} = ReLU (MLP (g_{j}^{t - 1}, g_{i}^{t} - g_{j}^{t - 1})),

(9)

Afterward, all intermediate features are agregated to generate the interpolated feature

h_{i}^{t}

as:

h_{i}^{t} = \sum_{j = 1}^{K} {\hat{w}}_{i j}^{t - 1, t} ⊙ v_{i j}^{t - 1, t} .

(10)

where ⊙ is an elemental multiplication.

As shown in Table 2, different variants of our dust filtering method can be built depending on the information (feature vector) used:

SI: Spatial + Intensity features.
STdm: Spatial + Temporal-magnitude-difference features.
STdv: Spatial + Temporal-vector-difference features.
STi: Spatial + Temporal-interpolated features.
SITdm: Spatial + Intensity + Temporal-magnitude-difference features.
SITdv: Spatial + Intensity + Temporal-vector-difference features.
SITi: Spatial + Intensity + Temporal-interpolated features.

Finally, in the case where the 3D data are acquired from a LIDAR mounted in a moving vehicle or robot, the odometry information is used to align the point clouds before the temporal features are computed. Thus, before calculating temporal features between point clouds

P^{t - 1}

and

P^{t}

, the points of

P^{t - 1}

are projected to t using the rotation and translation matrices between

t - 1

and t,

R_{t}

and

T_{t}

, respectively. These matrices are calculated from the vehicle’s odometry.

4. Experimental Results

4.1. UCHILE-Dust Database

The UCHILE-Dust dataset consists of real-world LIDAR point clouds captured using an OS0 sensor in various indoor and outdoor environments under dusty conditions. Each recording spans 15 seconds to 2 minutes, during which dust is actively dispersed in the scene. Each LIDAR frame is treated as an individual point cloud.

Recordings were saved either as PCAP files (direct OS0 stream) or ROS bag files (robot-mounted OS0 with odometry). A multi-step preprocessing pipeline was applied, including frame extraction, odometry alignment (moving sensor case), multi-echo merging, distance filtering, pre-labeling with dust-free references, manual annotation using labelCloud, and final conversion to the S3DIS (Stanford Large-Scale 3D Indoor Spaces) format. Table 3 shows a general overview of the database, considering the different subsets captured in different environments. It is important to note that the percentage of dust points in all subsets is less than 12%.

Interior 1 & 2 Subsets

Captured indoors at the Field Robotics Laboratory of the Advanced Mining Technology Center (AMTC) of the Universidad de Chile (UCHILE) using a static OS0 LiDAR sensor. Dust was manually dispersed across the scene, which includes a combination of glass and concrete surfaces that introduce complexities with transparent and reflective materials. A rock breaker hammer is present in the center of the room.

Exterior 1 & 2 Subsets

Outdoor captures in the AMTC courtyard. Dust was dispersed either between the sensor and a nearby wall or in open space to assess the impact of multi-path reflections.

Carén Subset

Captured in a large, dry, and windy open field with flat terrain and a quarry. The site is located in the Metropolitan Region of Chile and belongs to the UCHILE. A Panther robot equipped with an OS0 LiDAR was used to collect data while in motion. Dust was introduced using an air blower. Carén is the most realistic subset for mobile perception tasks.

Example point cloud frames for each subset are shown in Figure 1.

4.2. Experimental Setup

Each variant of the model (SI, STdm, STdv, STi, SITdm, SITdv, and SITi) was trained using its respective training and validation sets. Three different learning rates were used for each variant: 0.01, 0.008, and 0.005. The learning rate that produced the best results was selected. This was done because the average accuracy varied depending on the learning rate used in each experiment. In all cases a batch size of 16 was used. Other hyperparameters used are the following:

The CrossEntropyLoss was used as a loss function but considering weights for the classes, since they are unbalanced. This weight consisted of the inverse of the proportion of each class within the corresponding training dataset.
The number of training epochs was set at 100, but Early Stopping was implemented with 10 epochs of patience relative to the average accuracy value in the validation set. This metric was chosen as it is invariant to class imbalance.
Data Augmentation methods were used: rotations, scaling, occlusion, and noise.
A dropout rate of 0.7 was used to reduce overfitting and was applied to the last convolution layer before the classification layers.

As a baseline, the LIDROR method was used. The LIDROD parameters were determined using grid search and the following ranges:

μ \in 90, 100, 110

,

R \in 0.001, 0.01

,

ϕ \in 0.01, 0.05, 0.1

, and

N \in 4, 6, 8

. Table 4 shows the parameters obtained. The parameter

α

is the angular resolution of the sensor; it depends on the LIDAR model and was set to 0.006134.

The following metrics were employed to assess the quality of the various methods employed for dust classification: accuracy, precision, recall and F1-score. Accuracy is the proportion of all classifications that are correct, precision is the proportion of dust classifications that are actually dust, recall is the proportion of actual dust samples that are correctly classified as dust, and F1-score is the harmonic mean of precision and recall.

4.3. Results in Real Environments with Static Sensors

The results of applying different dust-filtering methods to the Interior 1, Interior 2, Exterior 1, and Exterior 2 datasets are shown in Table 5, Table 6, Table 7 and Table 8, respectively. In each instance, the methods are trained using exclusively data from the corresponding datasets. Table 2 presents the features utilized by the variants of the proposed neural-based method, while Table 4 shows the parameters of the LIDROR baseline method.

The results obtained allow for the following conclusions to be drawn. All variants of the proposed method demonstrate superior performance compared to the baseline, with the exception of SITdv when applied to the Interior 1 dataset. In the Interior 2 and Exterior 1 datasets, at least one of the proposed methods achieves an acceptable precision value (greater than or equal to 0.68).

The most challenging datasets are Interior 1 and Exterior 2. In these datasets, all methods have low precision. Given that in deployment in a real environment false positives can be costly, this dataset represents an ideal benchmark for further experimentation. Furthermore, low F1-Scores are expected in these kinds of datasets and are driven by the large imbalance between classes.

Finally, a comparison of the proposed variants reveals that the utilization of both temporal difference and intensity features is generally superior to the application of either temporal differences or intensity features alone.

4.4. Results in Real Environments with Moving Sensors

The results of applying different dust-filtering methods to the Carén dataset, in which the LIDAR sensor is mounted on a mobile robot, are shown in Table 9 and Table 10. Table 9 shows the results when the odometry correction is not used, and Table 10 shows the results when the odometry correction is used.

The results obtained allow for the following conclusions to be drawn. All variants of the proposed method largely outperform the baseline. In fact, LIDROR achieves very low precision with this dataset. Secondly, most of the results obtained by the proposed variants improve when odometry correction is used. The only two cases in which the accuracy decreases are those corresponding to the STi and STdv variants. The best results are obtained by the SITdv variant, with a precision value of 0.7, a recall of 0.97, and an F1-score of 0.82.

4.5. Measuring the Generalization Capabilities of the Method

To measure the generalization of the variants of the proposed methods, all methods were trained and validated using the Interior 1, Interior 2, Exterior 1, and Exterior 2 datasets but tested using the Carén dataset. The results of these experiments are shown in Table 13 and Table 12 for cases with and without odometry.

Upon comparing the results of the methods trained and validated using the Carén dataset (Table 9 and Table 10) reveals that there is only a slight advantage to in-domain training relative to training in one environment and testing in another (Table 13 and Table 12). This finding indicates that the model is learning underlying dust features rather than environment-specific features, such as the relative position of the objects in the scene. This finding suggests that the model is generalizing correctly.

A second conclusion is that, under this experimental setup, the benefits of using odometry correction are lost; in most cases, no improvement in results is observed. One possible reason is that the training and validation sets do not consider the correction of odometry.

Table 11. Generalization results. Training in AMTC datasets, testing in Carén dataset. No odometry correction

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.005	0.98	0.58	0.98	0.73
STdm	0.008	0.96	0.46	0.95	0.62
STdv	0.005	0.95	0.41	0.95	0.57
STi	0.005	0.96	0.64	0.94	0.76
SITdm	0.005	0.98	0.60	0.98	0.74
SITdv	0.008	0.97	0.54	0.98	0.70
SITi	0.01	0.98	0.57	0.98	0.72

Table 12. Generalization results. Training in AMTC datasets, testing in Carén dataset, with odometry correction

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.008	0.98	0.56	0.98	0.71
STdm	0.008	0.96	0.50	0.96	0.66
STdv	0.005	0.96	0.49	0.95	0.65
STi	0.01	0.96	0.52	0.95	0.67
SITdm	0.005	0.98	0.60	0.98	0.75
SITdv	0.008	0.98	0.58	0.98	0.73
SITi	0.008	0.98	0.57	0.98	0.72

4.6. Performance Comparison between PointNet++ and reduced-PointNet++

As a final experiment, a comparison was carried out between using the original PointNet++ architecture and the reduced PointNet++ architecture proposed here, for all of the methods presented in this work.

All experiments were carried out on a platform with a 12 GB VRAM NVIDIA GTX TITAN GPU, a 12-core Intel® Core™ i7-8700K processor running at 3.70 GHz, and 16 GB of RAM. Table 13 shows a comparison of the execution times of both architectures. It can be seen that the reduced PointNet++ architecture outperforms the original PointNet++ architecture in terms of speed. The execution times are, on average, 50% shorter.

It must be noted that this reduction in execution time does not sacrifice performance. Table 14 shows the results of using the original PointNet++ architecture when the methods are trained on the Interior 1, Interior 2, Exterior 1, and Exterior 2 datasets but tested on the Carén dataset, without using odometry information. These results are directly comparable to those in Table 13. Comparing both tables reveals that the results are similar. In some cases, the original PointNet++ obtains slightly better results, while in others, the proposed reduced PointNet++ obtains slightly better results.

These results show that reducing the network size is a valid approach for the dust detection problem, but it also shows that there is still room for further prunning since the reduction in parameters did not affect performance.

Table 13. Comparison of the execution times of PointNet++ versus reduced-PointNet++.

Method	PointNet++ (s)	reduced-PointNet++ (s)
SI	0.1383	0.0676
STdm	0.1398	0.0672
STdv	0.1419	0.0679
STI	0.1420	0.0694
SITdm	0.1372	0.0710
SITdv	0.1435	0.0664
SITi	0.1435	0.0630
Average	0.1409	0.0675

Table 14. Generalization results obtained using the original PointNet++ architecture. Training in AMTC datasets, testing in Carén dataset. No odometry correction.

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.005	0.98	0.55	0.98	0.71
STdm	0.005	0.96	0.46	0.96	0.62
STdv	0.005	0.95	0.51	0.94	0.66
STi	0.01	0.96	0.51	0.96	0.67
SITdm	0.005	0.98	0.58	0.98	0.73
SITdv	0.005	0.98	0.62	0.97	0.76
SITi	0.008	0.98	0.61	0.97	0.75

5. Discussion

This work presents a real-time dust filtering method for LIDAR point clouds based on a reduced PointNet++ architecture. From the results obtained, it is clear that the proposed approach outperforms the baseline heuristic method. Furthermore, the experiments demonstrated that temporal information is useful in distinguishing dust from static objects, and the integration of odometry further improved the performance in mobile applications.

This is a promising approach for robotic applications in challenging dust-filled environments, since it allows to filter dust from LiDAR measurements in order to perform basic navigation tasks such as SLAM, object avoidance, and emergency stops. Furthermore, the proposed reduction of the network size, which in turn reduces the computational requirements and inference times, allows the network to be deployed in embedded systems such as those found in robotic platforms, which are usually constrained by restrictions such as size and, for dusty environments, passive cooling.

The appoach was tested and developed using the UCHILE-Dust dataset, which will be made publicly available to support the development and benchmarking of future methods. The dataset includes indoor, outdoor, and mobile robot recordings, covering a wide range of scenarios with varying dust densities, allowing for a wide range of possible applications. Our experiments showed that models trained on UCHILE-Dust generalized well from static environments to mobile robot deployments. This suggests that models trained on this dataset learn dust-specific features rather than environment-specific features such as the relative positions of the objects in the scene (layout).

Author Contributions

Conceptualization, B.C., N.C. and J.R.; methodology, B.C. and N.C; software, B.C. and N.C; validation, B.C., N.C. and J.R.; formal analysis, B.C., N.C. and J.R.; investigation, B.C., N.C. and J.R.; resources, J.R.; data curation, B.C., N.C.; writing—original draft preparation, B.C., N.C. and J.R.; writing—review and editing, B.C., N.C. and J.R.; visualization, B.C. and N.C; supervision, N.C. and J.R.; project administration, J.R.; funding acquisition, J.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Research and Development Agency (ANID) of Chile through project grant AFB230001 for the Advanced Mining Technology Center (AMTC).

Conflicts of Interest

The authors declare no conflicts of interest.

References

Heinzler, R., Piewak, F., Schindler, P., y Stork, W., “Cnn-based lidar point cloud de- noising in adverse weather”, IEEE Robotics and Automation Letters, vol. 5, p. 2514– 2521, apr 2020. [CrossRef]
Stanislas, L., Suenderhauf, N., y Peynot, T., “Lidar-based detection of airborne particles for robust robot perception”, en Proceedings of the Australasian Conference on Robotics and Automation (ACRA) 2018 (Woodhead, I., ed.), Australasian Conference on Robo- tics and Automation, ACRA, pp. 1–8, Australia: Australian Robotics and Automation Association (ARAA), 2018, https://eprints.qut.edu.au/125130/.
Afzalaghaeinaeini, A., Design of Dust-Filtering Algorithms for LiDAR Sensors in Off- Road Vehicles Using the AI and Non-AI Methods. Tesis PhD, University of Ontario Institute of Technology, 2022.
Parsons, T., Seo, J., Kim, B., Lee, H., Kim, J.-C., y Cha, M., “Dust de-filtering in lidar applications with conventional and cnn filtering methods”, IEEE Access, vol. PP, pp. 1–1, 01 2024. [CrossRef]
Phillips, T., Guenther, N., y Mcaree, P., “When the dust settles: The four behaviors of lidar in the presence of fine airborne particulates”, Journal of Field Robotics, vol. 34, 02 2017. [CrossRef]
Charron, N., Phillips, S., y Waslander, S. L., “De-noising of lidar point clouds corrupted by snowfall”, en 2018 15th Conference on Computer and Robot Vision (CRV), pp. 254–261, 2018. [CrossRef]
Park, J.-I., Park, J., y Kim, K.-S., “Fast and accurate desnowing algorithm for lidar point clouds”, IEEE Access, vol. 8, pp. 160202–160212, 2020. [CrossRef]
Afzalaghaeinaeini, A., Seo, J., Lee, D., y Lee, H., “Design of dust-filtering algorithms for lidar sensors using intensity and range information in off-road vehicles”, Sensors, vol. 22, no. 11, 2022. [CrossRef]
Stanislas, L., Nubert, J., Dugas, D., Nitsch, J., Sünderhauf, N., Siegwart, R., Cadena, C., y Peynot, T., Airborne Particle Classification in LiDAR Point Clouds Using Deep Learning,pp.395–410. Springer, Singapore, 022021. [CrossRef]
Qi, C. R., Su, H., Mo, K., y Guibas, L. J., “Pointnet: Deep learning on point sets for 3d classification and segmentation”, en Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 652–660, 2017.
Qi,C.R.,Yi,L.,Su,H.,yGuibas,L.J.,“Pointnet++:Deephierarchicalfeaturelearning on point sets in a metric space”, Advances in neural information processing systems, vol. 30, 2017.
Shi, H., Wei, J., Wang, H., Liu, F., y Lin, G., “Learning temporal variations for 4d point cloud segmentation”, International Journal of Computer Vision, vol. 132, no. 12, pp. 5603–5617, 2024. [CrossRef]

Figure 1. Example frames from the UCHILE-Dust dataset from Exterior (left), Carén (middle), and Interior (right) subsets.

Table 1. Comparison of key blocks of PointNet++ and reduced-PointNet++, highlighting the modifications in the number of sampled points and the configuration of the MLPs.

Block	PointNet++	reduced-PointNet++
`SA1`	1024 pts, r=0.1, 32 nbr, MLP: [32, 32, 64]	512 pts, r=0.1, 32 nbr, MLP: [16, 16, 32]
`SA2`	256 pts, r=0.2, 32 nbr, MLP: [64, 64, 128]	128 pts, r=0.2, 32 nbr, MLP: [32, 32, 64]
`SA3`	64 pts, r=0.4, 32 nbr, MLP: [128, 128, 256]	-
`SA4`	16 pts, r=0.8, 32 nbr, MLP: [256, 256, 512]	-
`FP4`	in_ch: 768, MLP: [256, 256]	-
`FP3`	in_ch: 384, MLP: [256, 256]	-
`FP2`	in_ch: 320, MLP: [256, 128]	in_ch: 96, MLP: [64, 32]
`FP1`	in_ch: 128, MLP: [128, 128, 128]	in_ch: (32 + num_feats), MLP: [32, 32]
Conv1d	128 → 128	32 → 32
BatchNorm1d	128	32
Dropout	p=0.7	p=0.7
Conv1d (final)	128 → num_classes	32 → num_classes

Table 2. Variants of the proposed dust filtering method and the features used in each case. For simplicity in the notation, the temporal indices t are omited.

Variant	Features Vector
SI	$[x_{i}, y_{i}, z_{i}, g_{i}]$
STdm	$[x_{i}, y_{i}, z_{i}, m_{i}]$
STdv	$[x_{i}, y_{i}, z_{i}, Δ c_{i}]$
STi	$[x_{i}, y_{i}, z_{i}, h_{i}]$
SITdm	$[x_{i}, y_{i}, z_{i}, g_{i}, m_{i}]$
SITdv	$[x_{i}, y_{i}, z_{i}, g_{i}, Δ c_{i}]$
SITi	$[x_{i}, y_{i}, z_{i}, g_{i}, h_{i}]$

Table 3. Overview of the UCHILE-Dust dataset subsets, including number of recordings, number of point clouds, total number of points, % of dust points, data format, and train/val/test split percentages.

Subset	Recordings	Point Clouds	Points	% Dust	Format	Train / Val / Test (%)
Interior 1	10	1,874	72,741,326	4.1%	PCAP	82 / 09 / 09
Interior 2	12	1,740	71,225,483	11.2%	PCAP	70 / 15 / 15
Exterior 1	10	1,529	45,311,889	3.2%	PCAP	84 / 08 / 08
Exterior 2	13	1,885	75,820,483	3.6%	PCAP	66 / 17 / 17
Carén	13	7,089	234,929,532	8.1%	Rosbag	46 / 26 / 28

Table 4. Parameters used for the LIDROR method.

Subset	$μ$	$R_{m i n}$	$ϕ$	N
Interior 1	90	0.001	0.01	4
Interior 2	90	0.001	0.01	4
Exterior 1	90	0.01	0.01	4
Exterior 2	90	0.01	0.01	4
Carén	90	0.001	0.01	4

Table 5. Results of dust-filtering using different methods in the Interior-1 dataset.

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.008	0.79	0.12	0.93	0.22
STdm	0.008	0.87	0.17	1.00	0.29
STdv	0.008	0.86	0.17	0.99	0.29
STI	0.01	0.89	0.26	0.91	0.41
SITdm	0.01	0.86	0.17	0.97	0.29
SITdv	0.005	0.77	0.13	0.85	0.23
SITi	0.005	0.80	0.15	0.85	0.26
LIDROR	-	0.76	0.14	0.77	0.24

Table 6. Results of dust-filtering using different methods in the Interior-2 dataset.

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.008	0.94	0.53	0.94	0.68
STdm	0.008	0.92	0.6	0.88	0.71
STdv	0.01	0.84	0.5	0.74	0.59
STI	0.008	0.87	0.55	0.8	0.65
SITdm	0.008	0.94	0.68	0.91	0.78
SITdv	0.008	0.95	0.66	0.93	0.77
SITi	0.008	0.94	0.6	0.92	0.73
LIDROR	-	0.86	0.21	0.96	0.35

Table 7. Results of dust-filtering using different methods in the Exterior-1 dataset.

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.01	0.99	0.48	0.99	0.65
STdm	0.005	0.99	0.35	0.98	0.51
STdv	0.005	0.99	0.27	0.99	0.42
STi	0.005	0.97	0.46	0.95	0.62
SITdm	0.01	0.99	0.68	0.99	0.81
SITdv	0.005	0.99	0.54	0.98	0.70
SITi	0.01	0.99	0.51	0.98	0.67
LIDROR	-	0.91	0.03	0.98	0.06

Table 8. Results of dust-filtering using different methods in the Exterior-2 dataset.

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.005	0.98	0.38	0.99	0.55
STdm	0.01	0.98	0.35	0.99	0.52
STdv	0.01	0.97	0.36	0.97	0.53
STi	0.01	0.94	0.2	0.97	0.34
SITdm	0.005	0.98	0.38	0.99	0.55
SITdv	0.005	0.98	0.39	0.99	0.57
SITi	0.008	0.98	0.44	0.99	0.61
LIDROR	-	0.88	0.09	0.98	0.16

Table 9. Results of dust-filtering using different methods in the Carén dataset. No odometry correction

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.005	0.98	0.61	0.98	0.75
STdm	0.01	0.95	0.43	0.96	0.60
STdv	0.01	0.95	0.47	0.94	0.62
STi	0.008	0.96	0.72	0.94	0.81
SITdm	0.01	0.98	0.6	0.98	0.74
SITdv	0.008	0.98	0.62	0.98	0.76
SITi	0.005	0.98	0.63	0.97	0.76
LIDROR	-	0.89	0.18	0.98	0.03

Table 10. Results of dust-filtering using different methods with odometry correction in the Carén dataset.

Method	Learn. Rate	Accuracy (avg)	Precision (dust)	Recall (dust)	F1-Score (dust)
SI	0.005	0.98	0.61	0.98	0.75
STdm	0.008	0.96	0.56	0.95	0.70
STdv	0.008	0.95	0.41	0.96	0.58
STi	0.005	0.96	0.56	0.95	0.71
SITdm	0.005	0.98	0.64	0.98	0.77
SITdv	0.005	0.98	0.7	0.97	0.82
SITi	0.008	0.98	0.62	0.98	0.76
LIDROR	-	0.89	0.18	0.98	0.3

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Dust Filtering in LIDAR Point Clouds Using Deep Learning for Mining Applications

Abstract

Keywords:

Subject:

1. Introduction

2. Related Work

2.1. Traditional Dust Filtering Methods

2.2. Machine Learning Based Dust Filtering Methods