TY - JOUR

T1 - Improving the classification of land use objects using dense connectivity of convolutional neural networks

AU - Gujrathi, A.

AU - Yang, C.

AU - Rottensteiner, F.

AU - Buddhiraju, K. M.

AU - Heipke, C.

N1 - Funding information: We thank the Landesamt für Geoinformation und Landes-vermessung Niedersachsen(LGLN), the Landesamt für Ver-messung und Geoinformation Schleswig Holstein (LVermGeo) and the Landesamt für innere Verwaltung Mecklenburg-Vorpommern (LaiV-MV) for providing the test data and for their support of this project. The first author is a Master’s student at Centre of Studies in Resources Engineering, Indian Institute of Technology Bombay and a Combined Study and Practice Stays for Engineers from Developing Countries (KOSPIE) Scholar, funded by Deutscher Akademischer Austauschdienst (DAAD), whose support is gratefully acknowledged. We thank the Landesamt fur Geoinformation und Landes-vermessung Niedersachsen(LGLN), the Landesamt fur Ver-messung und Geoinformation Schleswig Holstein (LVermGeo) and the Landesamt fur innere Verwaltung Mecklenburg-Vorpommern (LaiV-MV) for providing the test data and for their support of this project. The first author is a Master s student at Centre of Studies in Resources Engineering, Indian Institute of Technology Bombay and a Combined Study and Practice Stays for Engineers from Developing Countries (KOSPIE) Scholar, funded by Deutscher Akademischer Austauschdienst (DAAD), whose support is gratefully acknowledged.

PY - 2020/8/12

Y1 - 2020/8/12

N2 - Land use is an important variable in remote sensing which describes the functions carried out on a piece of land in order to obtain benefits and is especially useful to the personnel working in the fields of urban management and planning. The land use information is maintained by national mapping agencies in geo-spatial databases. Commonly, land use data is stored in the form of polygon objects; the label of the object indicates land use. The main goal of classification of land use objects is to update an existing database in an automatic process. Recently, Convolutional Neural Networks (CNN) have been widely used to tackle this task utilizing high resolution aerial images (and derived data such as digital surface model). One big challenge classifying polygons is to deal with the large variation in their geometrical extent. For this challenge, we adopt the method of Yang et al. (2019) to decompose polygons into regular patches of fixed size. The decomposition leads to two sets of polygons: small and large, where the former suffers from a lower identification rate. In this paper, we propose CNN methods which incorporate dense connectivity and integrate it with intermediate information via global average pooling to improve land use classification, mainly focusing on small polygons. We present different network variants by incorporating intermediate information via global average pooling from different stages of the network. We test our methods on two sites; our experiments show that the dense connectivity and integration of intermediate information has a positive effect not only on the classification accuracy on the whole but also on the identification of small polygons.

AB - Land use is an important variable in remote sensing which describes the functions carried out on a piece of land in order to obtain benefits and is especially useful to the personnel working in the fields of urban management and planning. The land use information is maintained by national mapping agencies in geo-spatial databases. Commonly, land use data is stored in the form of polygon objects; the label of the object indicates land use. The main goal of classification of land use objects is to update an existing database in an automatic process. Recently, Convolutional Neural Networks (CNN) have been widely used to tackle this task utilizing high resolution aerial images (and derived data such as digital surface model). One big challenge classifying polygons is to deal with the large variation in their geometrical extent. For this challenge, we adopt the method of Yang et al. (2019) to decompose polygons into regular patches of fixed size. The decomposition leads to two sets of polygons: small and large, where the former suffers from a lower identification rate. In this paper, we propose CNN methods which incorporate dense connectivity and integrate it with intermediate information via global average pooling to improve land use classification, mainly focusing on small polygons. We present different network variants by incorporating intermediate information via global average pooling from different stages of the network. We test our methods on two sites; our experiments show that the dense connectivity and integration of intermediate information has a positive effect not only on the classification accuracy on the whole but also on the identification of small polygons.

KW - CNN

KW - DenseNet

KW - Geospatial land use database

KW - global average pooling

KW - Land use classification

UR - http://www.scopus.com/inward/record.url?scp=85091059067&partnerID=8YFLogxK

U2 - 10.5194/isprs-archives-XLIII-B2-2020-667-2020

DO - 10.5194/isprs-archives-XLIII-B2-2020-667-2020

M3 - Conference article

AN - SCOPUS:85091059067

VL - 43

SP - 667

EP - 673

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

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

SN - 1682-1750

IS - B2

T2 - 2020 24th ISPRS Congress - Technical Commission II

Y2 - 31 August 2020 through 2 September 2020

ER -