scholarly journals Semantic Segmentation of Tree-Canopy in Urban Environment with Pixel-Wise Deep Learning

2021 ◽  
Vol 13 (16) ◽  
pp. 3054
Author(s):  
José Augusto Correa Martins ◽  
Keiller Nogueira ◽  
Lucas Prado Osco ◽  
Felipe David Georges Gomes ◽  
Danielle Elis Garcia Furuya ◽  
...  
Keyword(s):  
Deep Learning ◽  
Urban Areas ◽  
Urban Forests ◽  
Semantic Segmentation ◽  
Urban Environments ◽  
Tree Canopy ◽  
Environmental Benefits ◽  
Urban Systems ◽  
Convolutional Network ◽  
Environmental Technologies

Urban forests are an important part of any city, given that they provide several environmental benefits, such as improving urban drainage, climate regulation, public health, biodiversity, and others. However, tree detection in cities is challenging, given the irregular shape, size, occlusion, and complexity of urban areas. With the advance of environmental technologies, deep learning segmentation mapping methods can map urban forests accurately. We applied a region-based CNN object instance segmentation algorithm for the semantic segmentation of tree canopies in urban environments based on aerial RGB imagery. To the best of our knowledge, no study investigated the performance of deep learning-based methods for segmentation tasks inside the Cerrado biome, specifically for urban tree segmentation. Five state-of-the-art architectures were evaluated, namely: Fully Convolutional Network; U-Net; SegNet; Dynamic Dilated Convolution Network and DeepLabV3+. The experimental analysis showed the effectiveness of these methods reporting results such as pixel accuracy of 96,35%, an average accuracy of 91.25%, F1-score of 91.40%, Kappa of 82.80% and IoU of 73.89%. We also determined the inference time needed per area, and the deep learning methods investigated after the training proved to be suitable to solve this task, providing fast and effective solutions with inference time varying from 0.042 to 0.153 minutes per hectare. We conclude that the semantic segmentation of trees inside urban environments is highly achievable with deep neural networks. This information could be of high importance to decision-making and may contribute to the management of urban systems. It should be also important to mention that the dataset used in this work is available on our website.

scholarly journals Semantic segmentation of PolSAR image data using advanced deep learning model

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Rajat Garg ◽  
Anil Kumar ◽  
Nikunj Bansal ◽  
Manish Prateek ◽  
Shashi Kumar
Keyword(s):  
Machine Learning ◽  
Remote Sensing ◽  
Deep Learning ◽  
Urban Area ◽  
Urban Areas ◽  
Learning Algorithms ◽  
Semantic Segmentation ◽  
Learning Model ◽  
Machine Learning Algorithms ◽  
Deep Learning Model

AbstractUrban area mapping is an important application of remote sensing which aims at both estimation and change in land cover under the urban area. A major challenge being faced while analyzing Synthetic Aperture Radar (SAR) based remote sensing data is that there is a lot of similarity between highly vegetated urban areas and oriented urban targets with that of actual vegetation. This similarity between some urban areas and vegetation leads to misclassification of the urban area into forest cover. The present work is a precursor study for the dual-frequency L and S-band NASA-ISRO Synthetic Aperture Radar (NISAR) mission and aims at minimizing the misclassification of such highly vegetated and oriented urban targets into vegetation class with the help of deep learning. In this study, three machine learning algorithms Random Forest (RF), K-Nearest Neighbour (KNN), and Support Vector Machine (SVM) have been implemented along with a deep learning model DeepLabv3+ for semantic segmentation of Polarimetric SAR (PolSAR) data. It is a general perception that a large dataset is required for the successful implementation of any deep learning model but in the field of SAR based remote sensing, a major issue is the unavailability of a large benchmark labeled dataset for the implementation of deep learning algorithms from scratch. In current work, it has been shown that a pre-trained deep learning model DeepLabv3+ outperforms the machine learning algorithms for land use and land cover (LULC) classification task even with a small dataset using transfer learning. The highest pixel accuracy of 87.78% and overall pixel accuracy of 85.65% have been achieved with DeepLabv3+ and Random Forest performs best among the machine learning algorithms with overall pixel accuracy of 77.91% while SVM and KNN trail with an overall accuracy of 77.01% and 76.47% respectively. The highest precision of 0.9228 is recorded for the urban class for semantic segmentation task with DeepLabv3+ while machine learning algorithms SVM and RF gave comparable results with a precision of 0.8977 and 0.8958 respectively.

Implementation of Deep Learning Neural Network for Retinal Images

2020 ◽  
pp. 77-95
Author(s):  
R. Murugan
Keyword(s):  
Deep Learning ◽  
Semantic Segmentation ◽  
Neural System ◽  
Neural Systems ◽  
Retinal Images ◽  
Manual Segmentation ◽  
Convolutional Network ◽  
The World ◽  
Deep Learning Neural Network ◽  
Wide Assortment

The retinal parts segmentation has been recognized as a key component in both ophthalmological and cardiovascular sickness analysis. The parts of retinal pictures, vessels, optic disc, and macula segmentations, will add to the indicative outcome. In any case, the manual segmentation of retinal parts is tedious and dreary work, and it additionally requires proficient aptitudes. This chapter proposes a supervised method to segment blood vessel utilizing deep learning methods. All the more explicitly, the proposed part has connected the completely convolutional network, which is normally used to perform semantic segmentation undertaking with exchange learning. The convolutional neural system has turned out to be an amazing asset for a few computer vision assignments. As of late, restorative picture investigation bunches over the world are rapidly entering this field and applying convolutional neural systems and other deep learning philosophies to a wide assortment of uses, and uncommon outcomes are rising constantly.

scholarly journals Transformer-Based Decoder Designs for Semantic Segmentation on Remotely Sensed Images

2021 ◽  
Vol 13 (24) ◽  
pp. 5100
Author(s):  
Teerapong Panboonyuen ◽  
Kulsawasd Jitkajornwanich ◽  
Siam Lawawirojwong ◽  
Panu Srestasathiern ◽  
Peerapon Vateekul
Keyword(s):  
Image Processing ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Language Processing ◽  
State Of The Art ◽  
Semantic Segmentation ◽  
Landsat 8 ◽  
Convolutional Network ◽  
Image Labeling ◽  
Feature Pyramid

Transformers have demonstrated remarkable accomplishments in several natural language processing (NLP) tasks as well as image processing tasks. Herein, we present a deep-learning (DL) model that is capable of improving the semantic segmentation network in two ways. First, utilizing the pre-training Swin Transformer (SwinTF) under Vision Transformer (ViT) as a backbone, the model weights downstream tasks by joining task layers upon the pretrained encoder. Secondly, decoder designs are applied to our DL network with three decoder designs, U-Net, pyramid scene parsing (PSP) network, and feature pyramid network (FPN), to perform pixel-level segmentation. The results are compared with other image labeling state of the art (SOTA) methods, such as global convolutional network (GCN) and ViT. Extensive experiments show that our Swin Transformer (SwinTF) with decoder designs reached a new state of the art on the Thailand Isan Landsat-8 corpus (89.8% F1 score), Thailand North Landsat-8 corpus (63.12% F1 score), and competitive results on ISPRS Vaihingen. Moreover, both our best-proposed methods (SwinTF-PSP and SwinTF-FPN) even outperformed SwinTF with supervised pre-training ViT on the ImageNet-1K in the Thailand, Landsat-8, and ISPRS Vaihingen corpora.

scholarly journals Prey abundance and urbanization influence the establishment of avian predators in a metropolitan landscape

2018 ◽  
Vol 285 (1890) ◽  
pp. 20182120 ◽  
Author(s):  
Jennifer D. McCabe ◽  
He Yin ◽  
Jennyffer Cruz ◽  
Volker Radeloff ◽  
Anna Pidgeon ◽  
...  
Keyword(s):  
Urban Areas ◽  
Wildlife Conservation ◽  
Temporal Dynamics ◽  
Prey Availability ◽  
Canopy Cover ◽  
Urban Environments ◽  
Tree Canopy ◽  
Population Declines ◽  
Natural Habitats ◽  
Tree Canopy Cover

Urbanization causes the simplification of natural habitats, resulting in animal communities dominated by exotic species with few top predators. In recent years, however, many predators such as hawks, and in the US coyotes and cougars, have become increasingly common in urban environments. Hawks in the Accipiter genus, especially, are recovering from widespread population declines and are increasingly common in urbanizing landscapes. Our goal was to identify factors that determine the occupancy, colonization and persistence of Accipiter hawks in a major metropolitan area. Through a novel combination of citizen science and advanced remote sensing, we quantified how urban features facilitate the dynamics and long-term establishment of Accipiter hawks. Based on data from Project FeederWatch, we quantified 21 years (1996–2016) of changes in the spatio-temporal dynamics of Accipiter hawks in Chicago, IL, USA. Using a multi-season occupancy model, we estimated Cooper's ( Accipiter cooperii ) and sharp-shinned ( A. striatus ) hawk occupancy dynamics as a function of tree canopy cover, impervious surface cover and prey availability. In the late 1990s, hawks occupied 26% of sites around Chicago, but after two decades, their occupancy fluctuated close to 67% of sites and they colonized increasingly urbanized areas. Once established, hawks persisted in areas with high levels of impervious surfaces as long as those areas supported high abundances of prey birds. Urban areas represent increasingly habitable environments for recovering predators, and understanding the precise urban features that drive colonization and persistence is important for wildlife conservation in an urbanizing world.

scholarly journals Strong variations in urban allergenicity riskscapes due to poor knowledge of tree pollen allergenic potential

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Rita Sousa-Silva ◽  
Audrey Smargiassi ◽  
Daniel Kneeshaw ◽  
Jérôme Dupras ◽  
Kate Zinszer ◽  
...  
Keyword(s):  
New York ◽  
Urban Areas ◽  
Urban Forest ◽  
Pollen Allergy ◽  
Urban Forests ◽  
Urban Environments ◽  
Tree Pollen ◽  
Allergenic Potential ◽  
Common Tree ◽  
Pollen Allergenicity

AbstractExposure to allergenic tree pollen is an increasing environmental health issue in urban areas. However, reliable, well-documented, peer-reviewed data on the allergenicity of pollen from common tree species in urban environments are lacking. Using the concept of ‘riskscape’, we present and discuss evidence on how different tree pollen allergenicity datasets shape the risk for pollen-allergy sufferers in five cities with different urban forests and population densities: Barcelona, Montreal, New York City, Paris, and Vancouver. We also evaluate how tree diversity can modify the allergenic risk of urban forests. We show that estimates of pollen exposure risk range from 1 to 74% for trees considered to be highly allergenic in the same city. This variation results from differences in the pollen allergenicity datasets, which become more pronounced when a city’s canopy is dominated by only a few species and genera. In an increasingly urbanized world, diverse urban forests offer a potentially safer strategy aimed at diluting sources of allergenic pollen until better allergenicity data is developed. Our findings highlight an urgent need for a science-based approach to guide public health and urban forest planning.

scholarly journals SEMANTIC SEGMENTATION OF AERIAL IMAGES IN URBAN AREAS WITH CLASS-SPECIFIC HIGHER-ORDER CLIQUES

2015 ◽  
Vol II-3/W4 ◽  
pp. 127-133 ◽  
Author(s):  
J. A. Montoya-Zegarra ◽  
J. D. Wegner ◽  
L. Ladický ◽  
K. Schindler
Keyword(s):  
Urban Areas ◽  
Semantic Segmentation ◽  
Higher Order ◽  
Urban Environments ◽  
Aerial Images ◽  
Shape Representations ◽  
Large Sets ◽  
High Level ◽  
Occurrence Patterns ◽  
Data Term

In this paper we propose an approach to multi-class semantic segmentation of urban areas in high-resolution aerial images with classspecific object priors for buildings and roads. What makes model design challenging are highly heterogeneous object appearances and shapes that call for priors beyond standard smoothness or co-occurrence assumptions. The data term of our energy function consists of a pixel-wise classifier that learns local co-occurrence patterns in urban environments. To specifically model the structure of roads and buildings, we add high-level shape representations for both classes by sampling large sets of putative object candidates. Buildings are represented by sets of compact polygons, while roads are modeled as a collection of long, narrow segments. To obtain the final pixel-wise labeling, we use a CRF with higher-order potentials that balances the data term with the object candidates. We achieve overall labeling accuracies of > 80%.

scholarly journals IMPROVING DEEP LEARNING BASED SEMANTIC SEGMENTATION WITH MULTI VIEW OUTLIER CORRECTION

2020 ◽  
Vol XLIII-B2-2020 ◽  
pp. 711-716
Author(s):  
T. Peters ◽  
C. Brenner ◽  
M. Song
Keyword(s):  
Deep Learning ◽  
Transfer Learning ◽  
Input Data ◽  
Semantic Segmentation ◽  
Object Point ◽  
Mobile Mapping ◽  
Convolutional Network ◽  
3D Object ◽  
The Mean ◽  
2D Images

Abstract. The goal of this paper is to use transfer learning for semi supervised semantic segmentation in 2D images: given a pretrained deep convolutional network (DCNN), our aim is to adapt it to a new camera-sensor system by enforcing predictions to be consistent for the same object in space. This is enabled by projecting 3D object points into multi-view 2D images. Since every 3D object point is usually mapped to a number of 2D images, each of which undergoes a pixelwise classification using the pretrained DCNN, we obtain a number of predictions (labels) for the same object point. This makes it possible to detect and correct outlier predictions. Ultimately, we retrain the DCNN on the corrected dataset in order to adapt the network to the new input data. We demonstrate the effectiveness of our approach on a mobile mapping dataset containing over 10’000 images and more than 1 billion 3D points. Moreover, we manually annotated a subset of the mobile mapping images and show that we were able to rise the mean intersection over union (mIoU) by approximately 10% with Deeplabv3+, using our approach.

scholarly journals Remote Sensing of Urban Forests

2019 ◽  
Vol 11 (20) ◽  
pp. 2383 ◽  
Author(s):  
Giovanni Sanesi ◽  
Vincenzo Giannico ◽  
Mario Elia ◽  
Raffaele Lafortezza
Keyword(s):  
Remote Sensing ◽  
Urban Areas ◽  
Spatial Dynamics ◽  
Urban Forests ◽  
Urban Environments ◽  
Special Issue ◽  
Wide Range ◽  
Temporal And Spatial ◽  
New Sensors

Urban forests and green infrastructures at large are of critical importance for contemporary cities as they provide a wide range of ecosystem services (ESS) that enhance the quality of life of urban dwellers. Remote sensing technologies have greatly contributed to assessing and mapping the spatial distribution of ESS in urban areas, although more research is needed given the availability of new sensors from multiple satellites and platforms and the particular characteristics of urban environments (e.g., high heterogeneity). This Special Issue hosts papers focusing on the temporal and spatial dynamics of urban forests with special attention given to the most recent remote sensing technologies as well as advanced methods for processing geospatial data and extracting meaningful information.

scholarly journals Deep Residual Autoencoder with Multiscaling for Semantic Segmentation of Land-Use Images

2019 ◽  
Vol 11 (18) ◽  
pp. 2142 ◽  
Author(s):  
Lianfa Li
Keyword(s):  
Land Use ◽  
Deep Learning ◽  
Semantic Segmentation ◽  
Remotely Sensed ◽  
Convolutional Network ◽  
Convolutional Networks ◽  
Residual Learning ◽  
Fully Convolutional Networks ◽  
Remotely Sensed Images ◽  
Real World Datasets

Semantic segmentation is a fundamental means of extracting information from remotely sensed images at the pixel level. Deep learning has enabled considerable improvements in efficiency and accuracy of semantic segmentation of general images. Typical models range from benchmarks such as fully convolutional networks, U-Net, Micro-Net, and dilated residual networks to the more recently developed DeepLab 3+. However, many of these models were originally developed for segmentation of general or medical images and videos, and are not directly relevant to remotely sensed images. The studies of deep learning for semantic segmentation of remotely sensed images are limited. This paper presents a novel flexible autoencoder-based architecture of deep learning that makes extensive use of residual learning and multiscaling for robust semantic segmentation of remotely sensed land-use images. In this architecture, a deep residual autoencoder is generalized to a fully convolutional network in which residual connections are implemented within and between all encoding and decoding layers. Compared with the concatenated shortcuts in U-Net, these residual connections reduce the number of trainable parameters and improve the learning efficiency by enabling extensive backpropagation of errors. In addition, resizing or atrous spatial pyramid pooling (ASPP) can be leveraged to capture multiscale information from the input images to enhance the robustness to scale variations. The residual learning and multiscaling strategies improve the trained model’s generalizability, as demonstrated in the semantic segmentation of land-use types in two real-world datasets of remotely sensed images. Compared with U-Net, the proposed method improves the Jaccard index (JI) or the mean intersection over union (MIoU) by 4-11% in the training phase and by 3-9% in the validation and testing phases. With its flexible deep learning architecture, the proposed approach can be easily applied for and transferred to semantic segmentation of land-use variables and other surface variables of remotely sensed images.

Sign in / Sign up

Export Citation Format

Share Document