Classification Methods for Remotely Sensed Data

Tso, Brandt; Mather, Paul M.

doi:10.4324/9780203303566

Cited by 411 publications

(406 citation statements)

References 0 publications

Supporting

Mentioning

378

Contrasting

Unclassified

Order By: Relevance

“…Finally, classification techniques reduce the range of values of the image (digital number) to another level (classes) through a system of allocation statistics. 9 Spectral classification methods can be differentiated into two groups: supervised and unsupervised classifications. Lerma 10 studied the application of photogrammetry and remote sensing using cameras sensitive to the visible and infrared spectra, obtaining rectified images classified by automated methods and thus determining the structural elements and pathologies detected on the façades of different buildings.…”

Section: Introductionmentioning

confidence: 99%

Recognition of materials and damage on historical buildings using digital image classification

Larriva¹,

Perea²,

Ureña³

et al. 2015

S. Afr. J. Sci.

View full text Add to dashboard Cite

Nowadays, techniques in digital image processing make it possible to detect damage, such as moisture or biological changes, on the surfaces of historical buildings. Digital classification techniques can be used to identify damages in construction materials in a non-destructive way. In this study, we evaluate the application of the object-oriented classification technique using photographs taken with a Fujifilm IS-Pro digital single lens reflex camera and the integration of the classified images in a three-dimensional model obtained through terrestrial laser scanning data in order to detect and locate damage affecting biocalcarenite stone employed in the construction of the Santa Marina Church (Córdoba, Spain). The Fujifilm IS-Pro camera captures spectral information in an extra-visible range, generating a wide spectral image with wavelengths ranging from ultraviolet to infrared. Techniques of object-oriented classification were applied, taking into account the shapes, textures, background information and spectral information in the image. This type of classification requires prior segmentation, defined as the search for homogeneous regions in an image. The second step is the classification process of these regions based on examples. The output data were classified according to the kind of damage that affects the biocalcarenite stone, reaching an overall classification accuracy of 92% and an excellent kappa statistic (85.7%). We have shown that multispectral classification with visible and near-infrared bands increased the degree of recognition among different damages. Post-analysis of these data integrated in a three-dimensional model allows us to obtain thematic maps with the size and position of the damage.

show abstract

Section: Introductionmentioning

confidence: 99%

Recognition of materials and damage on historical buildings using digital image classification

Larriva¹,

Perea²,

Ureña³

et al. 2015

S. Afr. J. Sci.

View full text Add to dashboard Cite

show abstract

“…The objective of Bayesian prediction is to achieve the maximum posterior probability by combining the prior and conditional probability distribution functions, so as to solve the classification problem effectively. The MRF classifier provides a convenient way to model the local properties of an image into positivity, Marknovianity and Homogeneity as its prior probability, together with the learnt likelihood from the MLP, which constitutes the MLP-MRF [47], [48]. Such local neighbourhood information can further be converted into its global equivalence of the Gibbs random field as an energy function based on the Hamersley-Clifford theorem [14].…”

Section: B Multilayer Perceptron Based Markov Random Field (Mlp-mrf)mentioning

confidence: 99%

VPRS-Based Regional Decision Fusion of CNN and MRF Classifications for Very Fine Resolution Remotely Sensed Images

Zhang

Sargent

Pan

et al. 2018

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

show abstract

“…2017, 9, x FOR PEER REVIEW 5 of 19 essential for the study of the main challenges of the environment at the current time, such as climate change, biodiversity, and the demand for agricultural and non-agricultural land uses in the face of world population growth. Of the different classification techniques that have been developed in the last decades, in the present work we have used the classification supervised by the maximum likelihood method [26,27]. From the two satellite images ( Figure 3) SENTINEL2A (MSI); dated 8 July 2017, and Landsat 8 (OLI) (Path: 199, Row: 34); acquisition date 27 July 2017, digital processing is performed using the Geographic Information System (GIS) software ArcGis v10.5.…”

Section: Remote Sensing Applied In the Delimitation Of Exposed Areasmentioning

confidence: 99%

“…It is also essential for the study of the main challenges of the environment at the current time, such as climate change, biodiversity, and the demand for agricultural and non-agricultural land uses in the face of world population growth. Of the different classification techniques that have been developed in the last decades, in the present work we have used the classification supervised by the maximum likelihood method [26,27].…”

Section: Remote Sensing Applied In the Delimitation Of Exposed Areasmentioning

confidence: 99%

Analysis of Flood Risk Due to Sea Level Rise in the Menor Sea (Murcia, Spain)

et al. 2018

View full text Add to dashboard Cite

This article analyzes the coastal vulnerability and flood risk due to sea level rise in the Menor Sea, Murcia (Spain). The vulnerability has been estimated from Sentinel-2 and Landsat 8 satellite imagery using Remote Sensing techniques. The risk of coastal flooding was calculated based on various time scenarios (X 0 -current, X 1 -100 years, X 2 -500 years, X 3 -1000 years, X 4 -Storm, X 5 -Tsunami). Geographic Information System and Remote Sensing techniques were used to build a regional model to predict changes in the mean sea level for several future scenarios, showing susceptible areas to be flooded. We have included new parameters to the model such as swell, mareal range or neotectonic factors aiming to better adjust it to the local conditions. The results showed a high risk of flooding in the barrier beach and coastal areas of the Menor Sea, with a medium to very high degree of vulnerability for the most populated and touristic areas. The maximum and minimum expected increase of the water sheet for the 100 year scenarios ranged from +4.22 to +5.69 m. This methodology can establish sectors that need structural measures to minimize the impact of the sea level rise occurring due to natural tendency in the short or long term, as well as by extreme events such as storm surges or tsunamis. Furthermore, it can be used in other areas to assist land management decision makers to reduce or mitigate the vulnerability and risk presented against the rise of the sea level.

show abstract

Classification Methods for Remotely Sensed Data

Cited by 411 publications

References 0 publications

Recognition of materials and damage on historical buildings using digital image classification

Recognition of materials and damage on historical buildings using digital image classification

VPRS-Based Regional Decision Fusion of CNN and MRF Classifications for Very Fine Resolution Remotely Sensed Images

Analysis of Flood Risk Due to Sea Level Rise in the Menor Sea (Murcia, Spain)

Contact Info

Product

Resources

About