Textural neural network and version space classifiers for remote sensing

Kaminsky, Edit J.; Barad, H.; Brown, Warick

doi:10.1080/014311697218737

Cited by 18 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, this work is often cited explicitly in relation to the specification of the target accuracy in many projects (e.g. Fisher and Langford 1996, Kaminsky et al 1997, Rogers et al 1997, Wright and Morrice 1997, Brown et al 2000, Franklin et al 2001, Lewis and Brown 2001, Carranza and Hale 2002, Yang and Lo 2002, Weng 2002, Rogan et al 2003, Shao et al 2003, Kerr and Cihlar 2004, Treitz and Rogan 2004, Mundia and Aniya 2005, Yang and Liu 2005. However, Anderson et al (1976) do not discuss the matter in great detail or set out to propose a universally adoptable set of map-evaluation criteria.…”

Section: Accuracy Targetmentioning

confidence: 99%

Harshness in image classification accuracy assessment

Foody

2008

International Journal of Remote Sensing

364

245

View full text Add to dashboard Cite

Thematic mapping via a classification analysis is one of the most common applications of remote sensing. The accuracy of image classifications is, however, often viewed negatively. Here, it is suggested that the approach to the evaluation of image classification accuracy typically adopted in remote sensing may often be unfair, commonly being rather harsh and misleading. It is stressed that the widely used target accuracy of 85% can be inappropriate and that the approach to accuracy assessment adopted commonly in remote sensing is pessimistically biased. Moreover, the maps produced by other communities, which are often used unquestioningly, may have a low accuracy if evaluated from the standard perspective adopted in remote sensing. A greater awareness of the problems encountered in accuracy assessment may help ensure that perceptions of classification accuracy are realistic and reduce unfair criticism of thematic maps derived from remote sensing.

show abstract

Section: Accuracy Targetmentioning

confidence: 99%

Harshness in image classification accuracy assessment

Foody

2008

International Journal of Remote Sensing

364

245

View full text Add to dashboard Cite

show abstract

“…Skidmore et al (1997) used the backpropagation learning algorithm with a MLP model and reported no statistically significant improvement of classification accuracy for mapping forest types. New architectures have been developed in an attempt to improve over MLP, like the fuzzy ARTMAP (Carpenter et al, 1992), textural neural networks (Kaminsky et al, 1997) and combinations of neural networks and expert systems (Murai and Omatu, 1997).…”

Section: Classifiersmentioning

confidence: 99%

Selection of imagery data and classifiers for mapping Brazilian semideciduous Atlantic forests

Carvalho

Clevers

Skidmore³

et al. 2004

International Journal of Applied Earth Observation and Geoinfor

View full text Add to dashboard Cite

“…They are 1) distribution-free (i.e., they do not require the data to conform to a statistical distribution known a priori) and 2) importance-free (i.e., they do not need information on the confidence level of each data source, which is reflected in the weights of the network after training [23]). On the other hand, the dependence of results on the shape and size of the processing window (which are usually fixed by the user on an a priori basis, i.e., these parameters are neither data-driven nor adaptive) is a well-known problem [19]. To avoid this dependence, a multichannel filtering approach, which is inherently multiresolution, is adopted before classification to provide a (nearly) orthogonal decomposition/reconstruction of the raw image [20]- [22].…”

Section: B Neural Networkmentioning

confidence: 99%

Contextual clustering for image labeling: an application to degraded forest assessment in Landsat TM images of the Brazilian Amazon

Sgrenzaroli¹,

Baraldi

Eva

et al. 2002

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

The Modified Adaptive Pappas Clustering (MPAC) algorithm, recently published in the image processing literature, is proposed as a valuable tool in the analysis of remotely sensed images where texture information is negligible. Owing to its contextual, adaptive, and multiresolutional labeling approach, MPAC preserves genuine but small regions, is easy to use (i.e., it requires minor user interaction to run), and is robust to changes in input parameters. As an application example, an MPAC-based three-stage classifier is applied to degraded forest detection in Landsat Thematic Mapper (TM) scenes of the Brazilian Amazon, where intermediate states of forest alterations caused by anthropogenic activities can be characterized by image structures 1-3 pixels wide. In three TM images of the Pará test site, where classification results are validated by means of qualitative and quantitative comparisons with aerial photos, degraded forest areas cover 13% to 45% of the image ground coverage. In the Mato Grosso test site, the degraded forest class overlaps with 1) 10% of the closed-canopy forest detected by the deforestation mapping program of the Food and Agriculture Organization (FAO, 1992), and 2) 19% of the closed-canopy forest detected by the Tropical Rain Forest Information Center (TRFIC, 1996). These figures are in line with the conclusions of a recent study where present estimates of annual deforestation for the Brazilian Amazon are speculated to capture less than half of the forest area that is actually impoverished each year.

show abstract

Textural neural network and version space classifiers for remote sensing

Cited by 18 publications

References 0 publications

Harshness in image classification accuracy assessment

Harshness in image classification accuracy assessment

Selection of imagery data and classifiers for mapping Brazilian semideciduous Atlantic forests

Contextual clustering for image labeling: an application to degraded forest assessment in Landsat TM images of the Brazilian Amazon

Contact Info

Product

Resources

About