scholarly journals Influence of sample distribution and prior probability adjustment on land cover classification

2016 ◽  
Author(s):  
D Pouliot ◽  
R Latifovic ◽  
W Parkinson
2021 ◽  
Vol 13 (22) ◽  
pp. 4594
Author(s):  
Chenxi Li ◽  
Zaiying Ma ◽  
Liuyue Wang ◽  
Weijian Yu ◽  
Donglin Tan ◽  
...  

High-quality training samples are essential for accurate land cover classification. Due to the difficulties in collecting a large number of training samples, it is of great significance to collect a high-quality sample dataset with a limited sample size but effective sample distribution. In this paper, we proposed an object-oriented sampling approach by segmenting image blocks expanded from systematically distributed seeds (object-oriented sampling approach) and carried out a rigorous comparison of seven sampling strategies, including random sampling, systematic sampling, stratified sampling (stratified sampling with the strata of land cover classes based on classification product, Latin hypercube sampling, and spatial Latin hypercube sampling), object-oriented sampling, and manual sampling, to explore the impact of training sample distribution on the accuracy of land cover classification when the samples are limited. Five study areas from different climate zones were selected along the China–Mongolia border. Our research identified the proposed object-oriented sampling approach as the first-choice sampling strategy in collecting training samples. This approach improved the diversity and completeness of the training sample set. Stratified sampling with strata defined by the combination of different attributes and stratified sampling with the strata of land cover classes had their limitations, and they performed well in specific situations when we have enough prior knowledge or high-accuracy product. Manual sampling was greatly influenced by the experience of interpreters. All these sampling strategies mentioned above outperformed random sampling and systematic sampling in this study. The results indicate that the sampling strategies of training datasets do have great impacts on the land cover classification accuracies when the sample size is limited. This paper will provide guidance for efficient training sample collection to increase classification accuracies.


2009 ◽  
Vol 15 (5) ◽  
pp. 16-23
Author(s):  
O.I. Sakhatsky ◽  
◽  
G.M. Zholobak ◽  
A.A. Makarova ◽  
O.A. Apostolov ◽  
...  

Author(s):  
Serge A. Wich ◽  
Lian Pin Koh

This chapter discusses how data that have been collected with drones can be used to derive orthomosaics and digital surface models through structure-from-motion software and how these can be processed further for land-cover classification or into vegetation metrics. Some examples of the various programs are provided as well. The chapter ends with a discussion on the approaches that have been used to automate counts of animals in drone images.


2021 ◽  
Vol 13 (6) ◽  
pp. 3070
Author(s):  
Patrycja Szarek-Iwaniuk

Urbanization processes are some of the key drivers of spatial changes which shape and influence land use and land cover. The aim of sustainable land use policies is to preserve and manage existing resources for present and future generations. Increasing access to information about land use and land cover has led to the emergence of new sources of data and various classification systems for evaluating land use and spatial changes. A single globally recognized land use classification system has not been developed to date, and various sources of land-use/land-cover data exist around the world. As a result, data from different systems may be difficult to interpret and evaluate in comparative analyses. The aims of this study were to compare land-use/land-cover data and selected land use classification systems, and to determine the influence of selected classification systems and spatial datasets on analyses of land-use structure in the examined area. The results of the study provide information about the existing land-use/land-cover databases, revealing that spatial databases and land use and land cover classification systems contain many equivalent land-use types, but also differ in various respects, such as the level of detail, data validity, availability, number of land-use types, and the applied nomenclature.


Sign in / Sign up

Export Citation Format

Share Document