Classification of Buildings and Vehicles in Google Map Satellite Images Using Random Forest Classifier

2020 ◽  
pp. 1073-1082
Author(s):  
Ritik Gautam ◽  
Saravanan Chandran ◽  
Jose Hormese
Keyword(s):  
Random Forest ◽  
Satellite Images ◽  
Random Forest Classifier

scholarly journals Deep Convolutional Neural Network for Mapping Smallholder Agriculture Using High Spatial Resolution Satellite Image

Sensors ◽  
2019 ◽  
Vol 19 (10) ◽  
pp. 2398 ◽  
Author(s):  
Bin Xie ◽  
Hankui K. Zhang ◽  
Jie Xue
Keyword(s):  
Neural Network ◽  
Random Forest ◽  
Land Cover ◽  
Convolutional Neural Network ◽  
Satellite Images ◽  
Spatial Information ◽  
Deep Convolutional Neural Network ◽  
Crop Types ◽  
Pooling Strategy

In classification of satellite images acquired over smallholder agricultural landscape with complex spectral profiles of various crop types, exploring image spatial information is important. The deep convolutional neural network (CNN), originally designed for natural image recognition in the computer vision field, can automatically explore high level spatial information and thus is promising for such tasks. This study tried to evaluate different CNN structures for classification of four smallholder agricultural landscapes in Heilongjiang, China using pan-sharpened 2 m GaoFen-1 (meaning high resolution in Chinese) satellite images. CNN with three pooling strategies: without pooling, with max pooling and with average pooling, were evaluated and compared with random forest. Two different numbers (~70,000 and ~290,000) of CNN learnable parameters were examined for each pooling strategy. The training and testing samples were systematically sampled from reference land cover maps to ensure sample distribution proportional to the reference land cover occurrence and included 60,000–400,000 pixels to ensure effective training. Testing sample classification results in the four study areas showed that the best pooling strategy was the average pooling CNN and that the CNN significantly outperformed random forest (2.4–3.3% higher overall accuracy and 0.05–0.24 higher kappa coefficient). Visual examination of CNN classification maps showed that CNN can discriminate better the spectrally similar crop types by effectively exploring spatial information. CNN was still significantly outperformed random forest using training samples that were evenly distributed among classes. Furthermore, future research to improve CNN performance was discussed.

scholarly journals Classification of EEG data for human mental state analysis using Random Forest Classifier

2018 ◽  
Vol 132 ◽  
pp. 1523-1532 ◽  
Author(s):  
Damodar Reddy Edla ◽  
Kunal Mangalorekar ◽  
Gauri Dhavalikar ◽  
Shubham Dodia
Keyword(s):  
Random Forest ◽  
Mental State ◽  
Random Forest Classifier ◽  
Eeg Data ◽  
State Analysis

Enhancing alpine glacial lakes detection and mapping using multi-source data and machine learning techniques

2020 ◽  
Author(s):  
Sonam Wangchuk ◽  
Tobias Bolch
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Satellite Images ◽  
Random Forest Classifier ◽  
Machine Learning Techniques ◽  
Glacial Lake ◽  
Glacial Lakes ◽  
Alpine Regions ◽  
Learning Techniques ◽  
Source Data

<p>An accurate detection and mapping of glacial lakes in the Alpine regions such as the Himalayas, the Alps and the Andes are challenged by many factors. These factors include 1) a small size of glacial lakes, 2) cloud cover in optical satellite images, 3) cast shadows from mountains and clouds, 4) seasonal snow in satellite images, 5) varying degree of turbidity amongst glacial lakes, and 6) frozen glacial lake surface. In our study, we propose a fully automated approach, that overcomes most of the above mentioned challenges, to detect and map glacial lakes accurately using multi-source data and machine learning techniques such as the random forest classifier algorithm. The multi-source data are from the Sentinel-1 Synthetic Aperture Radar data (radar backscatter), the Sentinel-2 multispectral instrument data (NDWI), and the SRTM digital elevation model (slope). We use these data as inputs for the rule-based segmentation of potential glacial lakes, where decision rules are implemented from the expert system. The potential glacial lake polygons are then classified either as glacial lakes or non-glacial lakes by the trained and tested random forest classifier algorithm. The performance of the method was assessed in eight test sites located across the Alpine regions (e.g. the Boshula mountain range and Koshi basin in the Himalayas, the Tajiks Pamirs, the Swiss Alps and the Peruvian Andes) of the word. We show that the proposed method performs efficiently irrespective of geographic, geologic, climatic, and glacial lake conditions.</p>

scholarly journals Assessment of Sentinel-2 Satellite Images and Random Forest Classifier for Rainforest Mapping in Gabon

Forests ◽  
2020 ◽  
Vol 11 (9) ◽  
pp. 941
Author(s):  
Adam Waśniewski ◽  
Agata Hościło ◽  
Bogdan Zagajewski ◽  
Dieudonné Moukétou-Tarazewicz
Keyword(s):  
Random Forest ◽  
Satellite Images ◽  
Forest Cover ◽  
Forest Type ◽  
Random Forest Classifier ◽  
Type Model ◽  
Spectral Bands ◽  
Freshwater Swamp ◽  
The Impact ◽  
Sentinel 2

This study is focused on the assessment of the potential of Sentinel-2 satellite images and the Random Forest classifier for mapping forest cover and forest types in northwest Gabon. The main goal was to investigate the impact of various spectral bands collected by the Sentinel-2 satellite, normalized difference vegetation index (NDVI) and digital elevation model (DEM), and their combination on the accuracy of the classification of forest cover and forest type. Within the study area, five classes of forest type were delineated: semi-evergreen moist forest, lowland forest, freshwater swamp forest, mangroves, and disturbed natural forest. The classification was performed using the Random Forest (RF) classifier. The overall accuracy for the forest cover ranged between 92.6% and 98.5%, whereas for forest type, the accuracy was 83.4 to 97.4%. The highest accuracy for forest cover and forest type classifications were obtained using a combination of spectral bands at spatial resolutions of 10 m and 20 m and DEM. In both cases, the use of the NDVI did not increase the classification accuracy. The DEM was shown to be the most important variable in distinguishing the forest type. Among the Sentinel-2 spectral bands, the red-edge followed by the SWIR contributed the most to the accuracy of the forest type classification. Additionally, the Random Forest model for forest cover classification was successfully transferred from one master image to other images. In contrast, the transferability of the forest type model was more complex, because of the heterogeneity of the forest type and environmental conditions across the study area.

Sign in / Sign up

Export Citation Format

Share Document