A review of supervised object-based land-cover image classification

Ma, Lei; Li, Manchun; Ma, Xiaowei; Cheng, Liang; Du, Peijun; Liu, Yongchao

doi:10.1016/j.isprsjprs.2017.06.001

Cited by 679 publications

(414 citation statements)

References 110 publications

Supporting

Mentioning

393

Contrasting

Unclassified

Order By: Relevance

“…In addition, the detection process is faster with the CNN-detector than with OBIA, which implies a higher user productivity. Our results also suggest that OBIA-methods and software could be further improved by including CNNs-classifiers ( [12]). This paper is organized as follows.…”

mentioning

confidence: 63%

“…Both of these datasets contain a large number of manually labeled images. For example, the Brazilian Coffee Scenes dataset contains 50,000 of 64 × 64-pixel tiles, labeled as coffee (1438) non-coffee (36,577) or mixed (12,989) and UC-Merced dataset contains 2100 256 × 256-pixel images labeled as belonging to 21 land-use classes, with 100 images corresponding to each class. Several works have reached classification accuracies greater than 95% on these database ( [22,29]).…”

Section: Land Cover Mappingmentioning

confidence: 99%

“…In this step, the user has to manually optimize and initialize a set of non-dimensional parameters namely: (i) The scale parameter, to control the average image segment size by defining the maximum allowed heterogeneity (in color and shape) for the resulting image objects. [12,38,39]). For this, the user has to introduce training sample sites for each class.…”

Section: Obia-based Detectionmentioning

confidence: 99%

“…To detect a specific object (e.g., a particular plant species individual) in an input image, first, the OBIA method divides the image into segments (e.g., by using a multi-resolution segmentation algorithm), and then classifies the segments based on their similarities (e.g., by using algorithms such as the k-nearest neighbor, Random forest, or Support vector machines [9][10][11][12]). This procedure has to be repeated and optimized for each single input image and the knowledge acquired (i.e., the OBIA segmentation and classification settings) from one input image cannot be directly reutilized in another.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Deep-learning Versus OBIA for Scattered Shrub Detection with Google Earth Imagery: Ziziphus lotus as Case Study

et al. 2017

View full text Add to dashboard Cite

There is a growing demand for accurate high-resolution land cover maps in many fields, e.g., in land-use planning and biodiversity conservation. Developing such maps has been traditionally performed using Object-Based Image Analysis (OBIA) methods, which usually reach good accuracies, but require a high human supervision and the best configuration for one image often cannot be extrapolated to a different image. Recently, deep learning Convolutional Neural Networks (CNNs) have shown outstanding results in object recognition in computer vision and are offering promising results in land cover mapping. This paper analyzes the potential of CNN-based methods for detection of plant species of conservation concern using free high-resolution Google Earth TM images and provides an objective comparison with the state-of-the-art OBIA-methods. We consider as case study the detection of Ziziphus lotus shrubs, which are protected as a priority habitat under the European Union Habitats Directive. Compared to the best performing OBIA-method, the best CNN-detector achieved up to 12% better precision, up to 30% better recall and up to 20% better balance between precision and recall. Besides, the knowledge that CNNs acquired in the first image can be re-utilized in other regions, which makes the detection process very fast. A natural conclusion of this work is that including CNN-models as classifiers, e.g., ResNet-classifier, could further improve OBIA methods. The provided methodology can be systematically reproduced for other species detection using our codes available through (https://github.com/EGuirado/CNN-remotesensing).

show abstract

mentioning

confidence: 63%

Section: Land Cover Mappingmentioning

confidence: 99%

Section: Obia-based Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep-learning Versus OBIA for Scattered Shrub Detection with Google Earth Imagery: Ziziphus lotus as Case Study

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Methods Thomas et al, 2003Lu et al, 2004Ma et al, 2017Maulik and Chakraborty, 2017Guisan and Thuiller, 2005Elith and Leathwick, 2009Franklin, 2010Li and Wang, 2013Morris et al, 2016Ashraf et al, 2017de Rivera and López-Quílez, 2017 can be approached from two perspectives: the data collection methods and the map production methods. Developments in data collection techniques in the last few decades have increased the types, amount and quality of data that can be collected for marine environmental characterization, particularly in terms of remotely sensed data (Brown et al, 2011;Kachelriess et al, 2014;Lecours et al, 2016b).…”

Section: Marine Habitat Mappingmentioning

confidence: 99%

On the Use of Maps and Models in Conservation and Resource Management (Warning: Results May Vary)

Lecours

2017

Front. Mar. Sci.

View full text Add to dashboard Cite

Conservation planning and management typically require accurate and spatially explicit data at scales that are relevant for conservation objectives. In marine conservation, these data are often combined with spatial analytical techniques to produce marine habitat maps. While marine habitat mapping is increasingly used to inform conservation efforts, this field is still relatively young and its methods are rapidly evolving. Because conservation efforts do not always specify standards or guidelines for the production of habitat maps, results can vary dramatically. As representations of real environmental characteristics, habitat maps are highly sensitive to how they are produced. In this review paper, I present four concepts that are known to cause variation in spatial representation and prediction of habitats: the methodology used, the quality and scale of the data, and the choice of variables in regards to fitness for use. I then discuss the potential antinomy associated with the use of habitat maps in conservation: while habitat maps have become an invaluable tool to inform and assist decision-making, maps of the same area built using different methods and data may provide dissimilar representations, thus providing different information and possibly leading to different decisions. Exploring the theories and methods that have proved effective in terrestrial conservation and the spatial sciences, and how they can be integrated in marine habitat mapping practices, could help improve maps used to support marine conservation efforts and result in more reliable products to inform conservation decisions. Having a strong, consistent, transparent, repeatable, and science-based protocol for data collection and mapping is essential for effectively supporting decision-makers in developing conservation and management plans. The development of user-friendly tools to assist in the application of such protocol is crucial to a widespread improvement in practices. I discuss the potential to use interactive and collaborative Geographic Information Systems (GIS) to encourage the conservation and management community, from data collectors and mapmakers to decision-makers, to move toward a digital resilience and to develop such science-based protocol. Until standards and protocols are developed, habitat maps should always be interpreted with care, and the methods and metadata associated with their production should always be explicitly stated.

show abstract

Applications of Multi‐Source and Multi‐Sensor Data Fusion of Remote Sensing for Forest Species Mapping

Mohammadpour¹,

Viegas²

2022

Advances in Remote Sensing for Forest Monitoring

View full text Add to dashboard Cite

A review of supervised object-based land-cover image classification

Cited by 679 publications

References 110 publications

Deep-learning Versus OBIA for Scattered Shrub Detection with Google Earth Imagery: Ziziphus lotus as Case Study

Deep-learning Versus OBIA for Scattered Shrub Detection with Google Earth Imagery: Ziziphus lotus as Case Study

On the Use of Maps and Models in Conservation and Resource Management (Warning: Results May Vary)

Applications of Multi‐Source and Multi‐Sensor Data Fusion of Remote Sensing for Forest Species Mapping

Contact Info

Product

Resources

About