Classification of Photogrammetric and Airborne LiDAR Point Clouds Using Machine Learning Algorithms

With the development of photogrammetry technologies, point clouds have found a wide range of use in academic and commercial areas. This situation has made it essential to extract information from point clouds. In particular, artificial intelligence applications have been used to extract information from point clouds to complex structures. Point cloud classification is also one of the leading areas where these applications are used. In this study, the classification of point clouds obtained by aerial photogrammetry and Light Detection and Ranging (LiDAR) technology belonging to the same region is performed by using machine learning. For this purpose, nine popular machine learning methods have been used. Geometric features obtained from point clouds were used for the feature spaces created for classification. Color information is also added to these in the photogrammetric point cloud. According to the LiDAR point cloud results, the highest overall accuracies were obtained as 0.96 with the Multilayer Perceptron (MLP) method. The lowest overall accuracies were obtained as 0.50 with the AdaBoost method. The method with the highest overall accuracy was achieved with the MLP (0.90) method. The lowest overall accuracy method is the GNB method with 0.25 overall accuracy.

Download Full-text

HIERARCHICAL HIGHER ORDER CRF FOR THE CLASSIFICATION OF AIRBORNE LIDAR POINT CLOUDS IN URBAN AREAS

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprsarchives-xli-b3-655-2016 ◽

2016 ◽

Vol XLI-B3 ◽

pp. 655-662 ◽

Cited By ~ 33

Author(s):

J. Niemeyer ◽

F. Rottensteiner ◽

U. Soergel ◽

C. Heipke

Keyword(s):

Urban Areas ◽

Point Cloud ◽

Conditional Random Field ◽

Point Clouds ◽

Higher Order ◽

Airborne Lidar ◽

Classification Result ◽

3D Lidar ◽

Point Level

We propose a novel hierarchical approach for the classification of airborne 3D lidar points. Spatial and semantic context is incorporated via a two-layer Conditional Random Field (CRF). The first layer operates on a point level and utilises higher order cliques. Segments are generated from the labelling obtained in this way. They are the entities of the second layer, which incorporates larger scale context. The classification result of the segments is introduced as an energy term for the next iteration of the point-based layer. This framework iterates and mutually propagates context to improve the classification results. Potentially wrong decisions can be revised at later stages. The output is a labelled point cloud as well as segments roughly corresponding to object instances. Moreover, we present two new contextual features for the segment classification: the <i>distance</i> and the <i>orientation of a segment with respect to the closest road</i>. It is shown that the classification benefits from these features. In our experiments the hierarchical framework improve the overall accuracies by 2.3% on a point-based level and by 3.0% on a segment-based level, respectively, compared to a purely point-based classification.

Download Full-text

Machine Learning-Based Supervised Classification of Point Clouds Using Multiscale Geometric Features

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi10030187 ◽

2021 ◽

Vol 10 (3) ◽

pp. 187

Author(s):

Muhammed Enes Atik ◽

Zaide Duran ◽

Dursun Zafer Seker

Keyword(s):

Machine Learning ◽

Point Cloud ◽

Supervised Classification ◽

Point Clouds ◽

Support Vector ◽

Geometric Features ◽

Mathematical Tool ◽

City Area ◽

3D Point Clouds

3D scene classification has become an important research field in photogrammetry, remote sensing, computer vision and robotics with the widespread usage of 3D point clouds. Point cloud classification, called semantic labeling, semantic segmentation, or semantic classification of point clouds is a challenging topic. Machine learning, on the other hand, is a powerful mathematical tool used to classify 3D point clouds whose content can be significantly complex. In this study, the classification performance of different machine learning algorithms in multiple scales was evaluated. The feature spaces of the points in the point cloud were created using the geometric features generated based on the eigenvalues of the covariance matrix. Eight supervised classification algorithms were tested in four different areas from three datasets (the Dublin City dataset, Vaihingen dataset and Oakland3D dataset). The algorithms were evaluated in terms of overall accuracy, precision, recall, F1 score and process time. The best overall results were obtained for four test areas with different algorithms. Dublin City Area 1 was obtained with Random Forest as 93.12%, Dublin City Area 2 was obtained with a Multilayer Perceptron algorithm as 92.78%, Vaihingen was obtained as 79.71% with Support Vector Machines and Oakland3D with Linear Discriminant Analysis as 97.30%.

Download Full-text

Classification of Mediterranean Shrub Species from UAV Point Clouds

Remote Sensing ◽

10.3390/rs14010199 ◽

2022 ◽

Vol 14 (1) ◽

pp. 199

Author(s):

Juan Pedro Carbonell-Rivera ◽

Jesús Torralba ◽

Javier Estornell ◽

Luis Ángel Ruiz ◽

Pablo Crespo-Peremarch

Keyword(s):

Forest Fires ◽

Point Clouds ◽

Gradient Boosting ◽

Mediterranean Forests ◽

Shrub Species ◽

Aerial Photogrammetry ◽

Wide Range ◽

Identification Of Species ◽

Behaviour Models

Modelling fire behaviour in forest fires is based on meteorological, topographical, and vegetation data, including species’ type. To accurately parameterise these models, an inventory of the area of analysis with the maximum spatial and temporal resolution is required. This study investigated the use of UAV-based digital aerial photogrammetry (UAV-DAP) point clouds to classify tree and shrub species in Mediterranean forests, and this information is key for the correct generation of wildfire models. In July 2020, two test sites located in the Natural Park of Sierra Calderona (eastern Spain) were analysed, registering 1036 vegetation individuals as reference data, corresponding to 11 shrub and one tree species. Meanwhile, photogrammetric flights were carried out over the test sites, using a UAV DJI Inspire 2 equipped with a Micasense RedEdge multispectral camera. Geometrical, spectral, and neighbour-based features were obtained from the resulting point cloud generated. Using these features, points belonging to tree and shrub species were classified using several machine learning methods, i.e., Decision Trees, Extra Trees, Gradient Boosting, Random Forest, and MultiLayer Perceptron. The best results were obtained using Gradient Boosting, with a mean cross-validation accuracy of 81.7% and 91.5% for test sites 1 and 2, respectively. Once the best classifier was selected, classified points were clustered based on their geometry and tested with evaluation data, and overall accuracies of 81.9% and 96.4% were obtained for test sites 1 and 2, respectively. Results showed that the use of UAV-DAP allows the classification of Mediterranean tree and shrub species. This technique opens a wide range of possibilities, including the identification of species as a first step for further extraction of structure and fuel variables as input for wildfire behaviour models.

Download Full-text

Mapping Canopy Heights in Dense Tropical Forests Using Low-Cost UAV-Derived Photogrammetric Point Clouds and Machine Learning Approaches

Remote Sensing ◽

10.3390/rs13183777 ◽

2021 ◽

Vol 13 (18) ◽

pp. 3777

Author(s):

He Zhang ◽

Marijn Bauters ◽

Pascal Boeckx ◽

Kristof Van Oost

Keyword(s):

Machine Learning ◽

Tropical Forests ◽

Expert Knowledge ◽

External Validation ◽

Point Clouds ◽

Machine Learning Algorithms ◽

Airborne Lidar ◽

Canopy Height ◽

Congo Basin ◽

Supervised Machine Learning

Tropical forests are a key component of the global carbon cycle and climate change mitigation. Field- or LiDAR-based approaches enable reliable measurements of the structure and above-ground biomass (AGB) of tropical forests. Data derived from digital aerial photogrammetry (DAP) on the unmanned aerial vehicle (UAV) platform offer several advantages over field- and LiDAR-based approaches in terms of scale and efficiency, and DAP has been presented as a viable and economical alternative in boreal or deciduous forests. However, detecting with DAP the ground in dense tropical forests, which is required for the estimation of canopy height, is currently considered highly challenging. To address this issue, we present a generally applicable method that is based on machine learning methods to identify the forest floor in DAP-derived point clouds of dense tropical forests. We capitalize on the DAP-derived high-resolution vertical forest structure to inform ground detection. We conducted UAV-DAP surveys combined with field inventories in the tropical forest of the Congo Basin. Using airborne LiDAR (ALS) for ground truthing, we present a canopy height model (CHM) generation workflow that constitutes the detection, classification and interpolation of ground points using a combination of local minima filters, supervised machine learning algorithms and TIN densification for classifying ground points using spectral and geometrical features from the UAV-based 3D data. We demonstrate that our DAP-based method provides estimates of tree heights that are identical to LiDAR-based approaches (conservatively estimated NSE = 0.88, RMSE = 1.6 m). An external validation shows that our method is capable of providing accurate and precise estimates of tree heights and AGB in dense tropical forests (DAP vs. field inventories of old forest: r2 = 0.913, RMSE = 31.93 Mg ha−1). Overall, this study demonstrates that the application of cheap and easily deployable UAV-DAP platforms can be deployed without expert knowledge to generate biophysical information and advance the study and monitoring of dense tropical forests.

Download Full-text

HIERARCHICAL HIGHER ORDER CRF FOR THE CLASSIFICATION OF AIRBORNE LIDAR POINT CLOUDS IN URBAN AREAS

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xli-b3-655-2016 ◽

2016 ◽

Vol XLI-B3 ◽

pp. 655-662 ◽

Cited By ~ 2

Author(s):

J. Niemeyer ◽

F. Rottensteiner ◽

U. Soergel ◽

C. Heipke

Keyword(s):

Urban Areas ◽

Point Cloud ◽

Conditional Random Field ◽

Point Clouds ◽

Higher Order ◽

Airborne Lidar ◽

Classification Result ◽

3D Lidar ◽

Point Level

Download Full-text

Classification of Brainwaves for Sleep Stages by High-Dimensional FFT Features from EEG Signals

Applied Sciences ◽

10.3390/app10051797 ◽

2020 ◽

Vol 10 (5) ◽

pp. 1797 ◽

Cited By ~ 2

Author(s):

Mera Kartika Delimayanti ◽

Bedy Purnama ◽

Ngoc Giang Nguyen ◽

Mohammad Reza Faisal ◽

Kunti Robiatul Mahmudah ◽

...

Keyword(s):

Machine Learning ◽

Sleep Stage ◽

Machine Learning Algorithms ◽

High Dimensional ◽

Sleep Stages ◽

Eeg Signals ◽

Stage Classification ◽

Sleep Stage Classification ◽

Low Dimensional

Manual classification of sleep stage is a time-consuming but necessary step in the diagnosis and treatment of sleep disorders, and its automation has been an area of active study. The previous works have shown that low dimensional fast Fourier transform (FFT) features and many machine learning algorithms have been applied. In this paper, we demonstrate utilization of features extracted from EEG signals via FFT to improve the performance of automated sleep stage classification through machine learning methods. Unlike previous works using FFT, we incorporated thousands of FFT features in order to classify the sleep stages into 2–6 classes. Using the expanded version of Sleep-EDF dataset with 61 recordings, our method outperformed other state-of-the art methods. This result indicates that high dimensional FFT features in combination with a simple feature selection is effective for the improvement of automated sleep stage classification.

Download Full-text

Deep Learning-Based Classification of Large-Scale Airborne LiDAR Point Cloud

Canadian Journal of Remote Sensing ◽

10.1080/07038992.2021.1927687 ◽

2021 ◽

pp. 1-15

Author(s):

Mathieu Turgeon-Pelchat ◽

Samuel Foucher ◽

Yacine Bouroubi

Keyword(s):

Deep Learning ◽

Point Cloud ◽

Large Scale ◽

Airborne Lidar

Download Full-text

A Machine Learning Approach to Study Glycosidase Activities from Bifidobacterium

Microorganisms ◽

10.3390/microorganisms9051034 ◽

2021 ◽

Vol 9 (5) ◽

pp. 1034

Author(s):

Carlos Sabater ◽

Lorena Ruiz ◽

Abelardo Margolles

Keyword(s):

Machine Learning ◽

Supervised Classification ◽

Machine Learning Algorithms ◽

Learning Approach ◽

Human Milk Oligosaccharides ◽

Future Studies ◽

High Fiber ◽

Machine Learning Approach ◽

Prebiotic Oligosaccharides

This study aimed to recover metagenome-assembled genomes (MAGs) from human fecal samples to characterize the glycosidase profiles of Bifidobacterium species exposed to different prebiotic oligosaccharides (galacto-oligosaccharides, fructo-oligosaccharides and human milk oligosaccharides, HMOs) as well as high-fiber diets. A total of 1806 MAGs were recovered from 487 infant and adult metagenomes. Unsupervised and supervised classification of glycosidases codified in MAGs using machine-learning algorithms allowed establishing characteristic hydrolytic profiles for B. adolescentis, B. bifidum, B. breve, B. longum and B. pseudocatenulatum, yielding classification rates above 90%. Glycosidase families GH5 44, GH32, and GH110 were characteristic of B. bifidum. The presence or absence of GH1, GH2, GH5 and GH20 was characteristic of B. adolescentis, B. breve and B. pseudocatenulatum, while families GH1 and GH30 were relevant in MAGs from B. longum. These characteristic profiles allowed discriminating bifidobacteria regardless of prebiotic exposure. Correlation analysis of glycosidase activities suggests strong associations between glycosidase families comprising HMOs-degrading enzymes, which are often found in MAGs from the same species. Mathematical models here proposed may contribute to a better understanding of the carbohydrate metabolism of some common bifidobacteria species and could be extrapolated to other microorganisms of interest in future studies.

Download Full-text

Delineating Smallholder Maize Farms from Sentinel-1 Coupled with Sentinel-2 Data Using Machine Learning

Sustainability ◽

10.3390/su13094728 ◽

2021 ◽

Vol 13 (9) ◽

pp. 4728

Author(s):

Zinhle Mashaba-Munghemezulu ◽

George Johannes Chirima ◽

Cilence Munghemezulu

Keyword(s):

Machine Learning ◽

Food Security ◽

Rural Communities ◽

Machine Learning Algorithms ◽

Support Vector ◽

Subsistence Agriculture ◽

Smallholder Farms ◽

Main Driver ◽

Sentinel 2

Rural communities rely on smallholder maize farms for subsistence agriculture, the main driver of local economic activity and food security. However, their planted area estimates are unknown in most developing countries. This study explores the use of Sentinel-1 and Sentinel-2 data to map smallholder maize farms. The random forest (RF), support vector (SVM) machine learning algorithms and model stacking (ST) were applied. Results show that the classification of combined Sentinel-1 and Sentinel-2 data improved the RF, SVM and ST algorithms by 24.2%, 8.7%, and 9.1%, respectively, compared to the classification of Sentinel-1 data individually. Similarities in the estimated areas (7001.35 ± 1.2 ha for RF, 7926.03 ± 0.7 ha for SVM and 7099.59 ± 0.8 ha for ST) show that machine learning can estimate smallholder maize areas with high accuracies. The study concludes that the single-date Sentinel-1 data were insufficient to map smallholder maize farms. However, single-date Sentinel-1 combined with Sentinel-2 data were sufficient in mapping smallholder farms. These results can be used to support the generation and validation of national crop statistics, thus contributing to food security.

Download Full-text

174 A comparison of machine learning algorithms in the classification of beef steers finished in feedlot

Journal of Animal Science ◽

10.1093/jas/skaa278.231 ◽

2020 ◽

Vol 98 (Supplement_4) ◽

pp. 126-127

Author(s):

Lucas S Lopes ◽

Christine F Baes ◽

Dan Tulpan ◽

Luis Artur Loyola Chardulo ◽

Otavio Machado Neto ◽

...

Keyword(s):

Machine Learning ◽

Decision Tree ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Final Decision ◽

Relevant Parameter ◽

Good Prediction ◽

Quality Traits ◽

C4.5 Decision Tree

Abstract The aim of this project is to compare some of the state-of-the-art machine learning algorithms on the classification of steers finished in feedlots based on performance, carcass and meat quality traits. The precise classification of animals allows for fast, real-time decision making in animal food industry, such as culling or retention of herd animals. Beef production presents high variability in its numerous carcass and beef quality traits. Machine learning algorithms and software provide an opportunity to evaluate the interactions between traits to better classify animals. Four different treatment levels of wet distiller’s grain were applied to 97 Angus-Nellore animals and used as features for the classification problem. The C4.5 decision tree, Naïve Bayes (NB), Random Forest (RF) and Multilayer Perceptron (MLP) Artificial Neural Network algorithms were used to predict and classify the animals based on recorded traits measurements, which include initial and final weights, sheer force and meat color. The top performing classifier was the C4.5 decision tree algorithm with a classification accuracy of 96.90%, while the RF, the MLP and NB classifiers had accuracies of 55.67%, 39.17% and 29.89% respectively. We observed that the final decision tree model constructed with C4.5 selected only the dry matter intake (DMI) feature as a differentiator. When DMI was removed, no other feature or combination of features was sufficiently strong to provide good prediction accuracies for any of the classifiers. We plan to investigate in a follow-up study on a significantly larger sample size, the reasons behind DMI being a more relevant parameter than the other measurements.

Download Full-text