scholarly journals SEMANTIC SEGMENTATION METHOD ACCELERATED QUANTITATIVE ANALYSIS OF THE SPATIAL CHARACTERISTICS OF TRADITIONAL VILLAGES

2021 ◽  
Vol XLVI-M-1-2021 ◽  
pp. 933-939
Author(s):  
M. Zhang ◽  
Z. Li ◽  
X. Wu
Keyword(s):  
Remote Sensing ◽  
Quantitative Analysis ◽  
Data Augmentation ◽  
Data Extraction ◽  
Remote Sensing Data ◽  
Semantic Segmentation ◽  
Mathematical Methods ◽  
Spatial Characteristics ◽  
Spatial Features ◽  
Short Period

Abstract. Rapid investigation and quantitative analysis are crucial for heritage conservation and renewal design. As an important category of architectural heritage - traditional settlements - with their large number and complex spatial characteristics, their spatial character patterns are an important support to assist settlement conservation and renewal design. However, the current means of analysis often requires manual data collection, secondary mapping of the collected data, extraction of individual elemental patterns and village boundaries. Then settlement boundary form, settlement density will be calculated by mathematical methods. The above methods are inefficient and prone to manual mapping errors, making it difficult to quantify and analyze a large number of traditional villages in a short period of time. Semantic segmentation is a computer vision technique for quickly segmenting different objects. Based on the collected remote sensing data of traditional villages, this paper established a dataset of semantic segmentation of spatial features of traditional settlements, segmenting village buildings, water systems, roads and plants. Using Transfer learning, data augmentation and other methods, a model was trained that can automatically segment elements of the villages. From the national traditional villages that have been announced so far, 60 traditional villages from different regions in the north and south were selected for analysis. The experiments show that the model established in this paper has an accuracy rate of above 86% in segmenting elements of villages, can effectively identify the location of different elements in remote sensing images, effectively improves the quantification rate of spatial features of settlements and saves the cost of mapping and data transcription. The results of the spatial characteristics of the 60 villages studied in this paper can also provide some theoretical basis and inspiration for the study, conservation, design and transformation of traditional villages.

scholarly journals Mapping the Unseen: Exploiting Super-Resolution for Semantic Segmentation in Low-Resolution Images

2020 ◽  
Author(s):  
Matheus B. Pereira ◽  
Jefersson Alex Dos Santos
Keyword(s):  
Remote Sensing ◽  
Pattern Recognition ◽  
Super Resolution ◽  
Remote Sensing Data ◽  
Semantic Segmentation ◽  
The Other ◽  
Aerial Imagery ◽  
Aerial Images ◽  
Remote Sensing Images ◽  
Low Resolution

High-resolution aerial images are usually not accessible or affordable. On the other hand, low-resolution remote sensing data is easily found in public open repositories. The problem is that the low-resolution representation can compromise pattern recognition algorithms, especially semantic segmentation. In this M.Sc. dissertation1 , we design two frameworks in order to evaluate the effectiveness of super-resolution in the semantic segmentation of low-resolution remote sensing images. We carried out an extensive set of experiments on different remote sensing datasets. The results show that super-resolution is effective to improve semantic segmentation performance on low-resolution aerial imagery, outperforming unsupervised interpolation and achieving semantic segmentation results comparable to highresolution data.

scholarly journals SDFCNv2: An Improved FCN Framework for Remote Sensing Images Semantic Segmentation

2021 ◽  
Vol 13 (23) ◽  
pp. 4902
Author(s):  
Guanzhou Chen ◽  
Xiaoliang Tan ◽  
Beibei Guo ◽  
Kun Zhu ◽  
Puyun Liao ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Receptive Field ◽  
Data Augmentation ◽  
Semantic Segmentation ◽  
Decision Fusion ◽  
Majority Voting ◽  
Model Parameters ◽  
Natural Scene ◽  
Segmentation Framework ◽  
Natural Scene Images

Semantic segmentation is a fundamental task in remote sensing image analysis (RSIA). Fully convolutional networks (FCNs) have achieved state-of-the-art performance in the task of semantic segmentation of natural scene images. However, due to distinctive differences between natural scene images and remotely-sensed (RS) images, FCN-based semantic segmentation methods from the field of computer vision cannot achieve promising performances on RS images without modifications. In previous work, we proposed an RS image semantic segmentation framework SDFCNv1, combined with a majority voting postprocessing method. Nevertheless, it still has some drawbacks, such as small receptive field and large number of parameters. In this paper, we propose an improved semantic segmentation framework SDFCNv2 based on SDFCNv1, to conduct optimal semantic segmentation on RS images. We first construct a novel FCN model with hybrid basic convolutional (HBC) blocks and spatial-channel-fusion squeeze-and-excitation (SCFSE) modules, which occupies a larger receptive field and fewer network model parameters. We also put forward a data augmentation method based on spectral-specific stochastic-gamma-transform-based (SSSGT-based) during the model training process to improve generalizability of our model. Besides, we design a mask-weighted voting decision fusion postprocessing algorithm for image segmentation on overlarge RS images. We conducted several comparative experiments on two public datasets and a real surveying and mapping dataset. Extensive experimental results demonstrate that compared with the SDFCNv1 framework, our SDFCNv2 framework can increase the mIoU metric by up to 5.22% while only using about half of parameters.

scholarly journals Assessment of Change in the Built-Up Index of Uyo Metropolis and Its Environs Using Remote Sensing

2019 ◽  
Vol 4 (6) ◽  
pp. 84-89 ◽  
Author(s):  
Aniekan Effiong Eyoh ◽  
Akwaowo Ekpa
Keyword(s):  
Remote Sensing ◽  
Land Use ◽  
Quantitative Analysis ◽  
Land Cover ◽  
Agricultural Land ◽  
Remote Sensing Data ◽  
Land Use Land Cover ◽  
Positive Growth ◽  
Bare Land ◽  
Erdas Imagine

The research was aim at assessing the change in the Built-up Index of Uyo metropolis and its environs from 1986 to 2018, using remote sensing data. To achieve this, a quantitative analysis of changes in land use/land cover within the study area was undertaken using remote sensing dataset of Landsat TM, ETM+ and OLI sensor images of 1986, 2000 and 2018 respectively. Supervised classification, using the maximum likelihood algorithm, was used to classify the study area into four major land use/land cover types; built-up land, bare land/agricultural land, primary swamp vegetation and secondary vegetation. Image processing was carried out using ERDAS IMAGINE and ArcGIS software. The Normalised Difference Built-up Index (NDBI) was calculated to obtain the built-up index for the study area in 1986, 2000 and 2018 as -0.20 to +0.45, -0.13 to +0.55 and -0.19 to +0.63 respectively. The result of the quantitative analysis of changes in land use/land cover indicated that Built-up Land had been on a constant and steady positive growth from 6.76% in 1986 to 11.29% in 2000 and 44.04% in 2018.

scholarly journals Determination of the Pyro-Factor Ecological Successions Duration in the Zonal Landscapes of the Volgograd Region According to the Remote Sensing Data

2020 ◽  
pp. 44-53
Author(s):  
Stanislav Shinkarenko ◽  
◽  
Asel Berdengalieva ◽  
Valeriya Doroshenko ◽  
Kseniya Oleynikova ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Spectral Characteristics ◽  
Remote Sensing Data ◽  
Field Studies ◽  
Spatial Characteristics ◽  
Sensing Data ◽  
Volgograd Region ◽  
Burnt Areas ◽  
Steppe Fires

The aim of the work is to determine the spatial characteristics of the distribution of the burnt areas of natural zonal landscapes of the Volgograd region with different duration of pyro-factor successions, taking into account the frequency of fires. Based on the previously developed thematic geo-information layers of the steppe fires in the region using overlay operations, the duration of post-pyrogenic periods in the municipal districts of the region was determined, taking into account the total number of fires from 1998–2018. The largest areas covered by fire have a succession duration of 2–3 years and 12–14 years at the beginning of 2019, which corresponds to the fires of 2016–2017 and 2005–2007, respectively. Large areas after the fires of 2001–2002 are located in Ilovlinsky, Kletsky, Pallasovsky and Surovikinsky districts. The largest area of land covered by fire in 2004–2006 is located in the Danilovsky, Ilovlinsky, Olkhovsky and Pallas districts. In our opinion, landscapes affected by fire no more than 5–7 years ago are suitable for the analysis of pyrogenic shifts. These territories are located in Frolovsky, Chernyshkovsky, Kotovsky, Ilovlinsky, Pallasovsky, Leninsky, Kamyshinsky, Staropoltavsky districts. The results will serve as the basis for field studies and the analysis of the spectral characteristics of overgrowing burns from remote sensing materials.

scholarly journals High Accuracy Geochemical Map Generation Method by a Spatial Autocorrelation-Based Mixture Interpolation Using Remote Sensing Data

2020 ◽  
Vol 12 (12) ◽  
pp. 1991
Author(s):  
Chenhui Huang ◽  
Akinobu Shibuya
Keyword(s):  
Remote Sensing ◽  
Spatial Autocorrelation ◽  
Data Augmentation ◽  
Prediction Models ◽  
Spatial Information ◽  
Thermal Emission ◽  
Remote Sensing Data ◽  
Regression Problem ◽  
Sensing Data ◽  
Multivariate Prediction

Generating a high-resolution whole-pixel geochemical contents map from a map with sparse distribution is a regression problem. Currently, multivariate prediction models like machine learning (ML) are constructed to raise the geoscience mapping resolution. Methods coupling the spatial autocorrelation into the ML model have been proposed for raising ML prediction accuracy. Previously proposed methods are needed for complicated modification in ML models. In this research, we propose a new algorithm called spatial autocorrelation-based mixture interpolation (SABAMIN), with which it is easier to merge spatial autocorrelation into a ML model only using a data augmentation strategy. To test the feasibility of this concept, remote sensing data including those from the advanced spaceborne thermal emission and reflection radiometer (ASTER), digital elevation model (DEM), and geophysics (geomagnetic) data were used for the feasibility study, along with copper geochemical and copper mine data from Arizona, USA. We explained why spatial information can be coupled into an ML model only by data augmentation, and introduced how to operate data augmentation in our case. Four tests—(i) cross-validation of measured data, (ii) the blind test, (iii) the temporal stability test, and (iv) the predictor importance test—were conducted to evaluate the model. As the results, the model’s accuracy was improved compared with a traditional ML model, and the reliability of the algorithm was confirmed. In summary, combining the univariate interpolation method with multivariate prediction with data augmentation proved effective for geological studies.

scholarly journals Multi-Object Segmentation in Complex Urban Scenes from High-Resolution Remote Sensing Data

2021 ◽  
Vol 13 (18) ◽  
pp. 3710
Author(s):  
Abolfazl Abdollahi ◽  
Biswajeet Pradhan ◽  
Nagesh Shukla ◽  
Subrata Chakraborty ◽  
Abdullah Alamri
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Urban Areas ◽  
Object Segmentation ◽  
Remote Sensing Data ◽  
Semantic Segmentation ◽  
Aerial Images ◽  
Urban Scenes ◽  
Sensing Data ◽  
Boundary Information

Terrestrial features extraction, such as roads and buildings from aerial images using an automatic system, has many usages in an extensive range of fields, including disaster management, change detection, land cover assessment, and urban planning. This task is commonly tough because of complex scenes, such as urban scenes, where buildings and road objects are surrounded by shadows, vehicles, trees, etc., which appear in heterogeneous forms with lower inter-class and higher intra-class contrasts. Moreover, such extraction is time-consuming and expensive to perform by human specialists manually. Deep convolutional models have displayed considerable performance for feature segmentation from remote sensing data in the recent years. However, for the large and continuous area of obstructions, most of these techniques still cannot detect road and building well. Hence, this work’s principal goal is to introduce two novel deep convolutional models based on UNet family for multi-object segmentation, such as roads and buildings from aerial imagery. We focused on buildings and road networks because these objects constitute a huge part of the urban areas. The presented models are called multi-level context gating UNet (MCG-UNet) and bi-directional ConvLSTM UNet model (BCL-UNet). The proposed methods have the same advantages as the UNet model, the mechanism of densely connected convolutions, bi-directional ConvLSTM, and squeeze and excitation module to produce the segmentation maps with a high resolution and maintain the boundary information even under complicated backgrounds. Additionally, we implemented a basic efficient loss function called boundary-aware loss (BAL) that allowed a network to concentrate on hard semantic segmentation regions, such as overlapping areas, small objects, sophisticated objects, and boundaries of objects, and produce high-quality segmentation maps. The presented networks were tested on the Massachusetts building and road datasets. The MCG-UNet improved the average F1 accuracy by 1.85%, and 1.19% and 6.67% and 5.11% compared with UNet and BCL-UNet for road and building extraction, respectively. Additionally, the presented MCG-UNet and BCL-UNet networks were compared with other state-of-the-art deep learning-based networks, and the results proved the superiority of the networks in multi-object segmentation tasks.

Sign in / Sign up

Export Citation Format

Share Document