scholarly journals CLASSIFICATION OF DATA FROM AIRBORNE LIDAR BATHYMETRY WITH RANDOM FOREST ALGORITHM BASED ON DIFFERENT FEATURE VECTORS

2019 ◽  
Vol XLII-2/W16 ◽  
pp. 143-148
Author(s):  
T. Kogut ◽  
M. Weistock ◽  
K. Bakuła
Keyword(s):  
Random Forest ◽  
Confusion Matrix ◽  
Airborne Lidar ◽  
Random Forest Algorithm ◽  
Data Set ◽  
Waveform Data ◽  
Feature Vectors ◽  
Full Waveform ◽  
Airborne Lidar Bathymetry

<p><strong>Abstract.</strong> Modern full-waveform laser bathymetric scanners offer the possibility of a practical application of airborne laser bathymetry (ALB) data algorithms as a valuable source of information in the study of the aquatic environment. The reliability of the obtained results and the efficiency of the classification depend on the applied features. The input data for the classifier should consist of variables that have the ability to discriminate within the data set, for the detection and classification of objects on the seabed. The automatic detection of underwater objects is based on machine learning solutions. In this paper, the ALB data were used to present a classification process based on the random forest algorithm. The classification was carried out using two independent approaches with two feature vectors. The quality of classifications based on the full-waveform features vector and the geometric features vector was compared. The efficiency of each classification was verified using a confusion matrix. The obtained efficiency of the point classification in both cases was about 100% for the water surface, 99.9% for the seabed and about 60% for underwater objects. Better results for the classification of objects were obtained for the features vector based on features obtained directly from full-waveform data than for the vector obtained from geometric relationships in the point cloud.</p>

scholarly journals WATER TURBIDITY ESTIMATION FROM LIDAR BATHYMETRY DATA BY FULL-WAVEFORM ANALYSIS – COMPARISON OF TWO APPROACHES

2021 ◽  
Vol XLIII-B2-2021 ◽  
pp. 681-688
Author(s):  
K. Richter ◽  
D. Mader ◽  
P. Westfeld ◽  
H.-G. Maas
Keyword(s):  
Measurement Data ◽  
Airborne Lidar ◽  
Waveform Analysis ◽  
Estimation Methods ◽  
Data Set ◽  
Water Turbidity ◽  
Waveform Data ◽  
Full Waveform ◽  
Novel Approach ◽  
Airborne Lidar Bathymetry

Abstract. Airborne LiDAR bathymetry is an efficient technique for surveying the bottom of shallow waters. In addition, the measurement data contain valuable information about the local turbidity conditions in the water body. The extraction of this information requires appropriate evaluation methods examining the decay of the recorded waveform signal. Existing approaches are based on several assumptions concerning the influence of the ALB system on the waveform signal, the extraction of the volume backscatter, and the directional independence of turbidity. The paper presents a novel approach that overcomes the existing limitations using two alternative turbidity estimation methods as well as different variants of further processed full-waveform data. For validation purposes, the approach was applied to a data set of a shallow inland water. The results of the quantitative evaluation show, which method and which data basis is best suited for the derivation of area wide water turbidity information.

scholarly journals Potential of a Non-linear Full-Waveform Stacking Technique in Airborne LiDAR Bathymetry

2021 ◽  
Author(s):  
David Mader ◽  
Katja Richter ◽  
Patrick Westfeld ◽  
Hans-Gerd Maas
Keyword(s):  
Water Depth ◽  
High Reliability ◽  
Bottom Topography ◽  
Airborne Lidar ◽  
Majority Voting ◽  
Waveform Data ◽  
Full Waveform ◽  
Airborne Lidar Bathymetry ◽  
Water Bottom

AbstractAirborne LiDAR bathymetry is an efficient measurement method for area-wide acquisition of water bottom topography in shallow water areas. However, the method has a limited penetration depth into water bodies due to water turbidity. This affects the accuracy and reliability of the determination of water bottom points in waters with high turbidity or larger water depths. Furthermore, the coverage of the water bottom topography is also limited. In this contribution, advanced processing methods are presented with the goal of increasing the evaluable water depth, resulting in an improved coverage of the water bottom by measurement points. The methodology moves away from isolated evaluation of individual signals to a determination of water bottom echoes, taking into account information from closely adjacent measurements, assuming that these have similar or correlated characteristics. The basic idea of the new processing approach is the combination of closely adjacent full-waveform data using full-waveform stacking techniques. In contrast to established waveform stacking techniques, we do not apply averaging, which entails low-pass filtering effects, but a modified majority voting technique. This has the effect of amplification of repeating weak characteristics and an improvement of the signal-noise-ratio. As a consequence, it is possible to detect water bottom points that cannot be detected by standard methods. The results confirm an increased penetration water depth by about 27% with a high reliability of the additionally extracted water bottom points along with a larger coverage of the water bottom topography.

scholarly journals Classification of Continuous Sky Brightness Data Using Random Forest

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Rhorom Priyatikanto ◽  
Lidia Mayangsari ◽  
Rudi A. Prihandoko ◽  
Agustinus G. Admiranto
Keyword(s):  
Random Forest ◽  
Observation Time ◽  
Classification Model ◽  
Posterior Distributions ◽  
Random Forest Algorithm ◽  
Negative Effect ◽  
Sky Brightness ◽  
Global Statistics ◽  
Effect Of Light

Sky brightness measuring and monitoring are required to mitigate the negative effect of light pollution as a byproduct of modern civilization. Good handling of a pile of sky brightness data includes evaluation and classification of the data according to its quality and characteristics such that further analysis and inference can be conducted properly. This study aims to develop a classification model based on Random Forest algorithm and to evaluate its performance. Using sky brightness data from 1250 nights with minute temporal resolution acquired at eight different stations in Indonesia, datasets consisting of 15 features were created to train and test the model. Those features were extracted from the observation time, the global statistics of nightly sky brightness, or the light curve characteristics. Among those features, 10 are considered to be the most important for the classification task. The model was trained to classify the data into six classes (1: peculiar data, 2: overcast, 3: cloudy, 4: clear, 5: moonlit-cloudy, and 6: moonlit-clear) and then tested to achieve high accuracy (92%) and scores (F-score = 84% and G-mean = 84%). Some misclassifications exist, but the classification results are considerably good as indicated by posterior distributions of the sky brightness as a function of classes. Data classified as class-4 have sharp distribution with typical full width at half maximum of 1.5 mag/arcsec2, while distributions of class-2 and -3 are left skewed with the latter having lighter tail. Due to the moonlight, distributions of class-5 and -6 data are more smeared or have larger spread. These results demonstrate that the established classification model is reasonably good and consistent.

Sign in / Sign up

Export Citation Format

Share Document