Aerial Image Texture Information in the Estimation of Northern Deciduous and Mixed Wood Forest Leaf Area Index (LAI)

Wulder, Michael A.; LeDrew, E.; Franklin, Steven E.; Lavigne, M. B.

doi:10.1016/s0034-4257(97)00169-7

Cited by 184 publications

(93 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In a similar manner, they have also been used as a tool to model the spatial variation within images (Woodcock et al 1988b, St-Onge and Cavayas 1995, Wulder et al 1998, Treitz 2001.…”

Section: Geostatisticsmentioning

confidence: 99%

“…Several studies have used geostatistical tools, specifically the variogram, to describe the magnitude of variation (sill) and the extent of spatial dependence (range) (Curran 1988, Woodcock et al 1988a. More recent research has found that range values are useful for determining the maximum size of processing windows or limits to spatial resolution (Cohen et al 1990, Franklin and McDermid 1993, Wulder et al 1998.…”

Section: Geostatisticsmentioning

confidence: 99%

“…Many studies have demonstrated the efficacy of this suite of quantitative techniques for estimating the optimum spatial resolution for remotely sensed data (Curran 1988, Atkinson 1993, and they have also been used as a tool to model the spatial variation within images (Woodcock et al 1988b, St-Onge and Cavayas 1995, Wulder et al 1996, 1998. In this study, the relationship between local image variance and semivariance was used to demonstrate that image variance, used in a multiscale approach, is an alternative measure of image texture.…”

Section: Introductionmentioning

confidence: 96%

“…Several studies have addressed this difficulty by calculating a single spatial variable and adding it to a classification procedure. Overall improvement in classification accuracy indicates that the addition of image texture improves image classification (Jensen 1982, Franklin and Peddle 1987, 1989, Cohen et al 1990, Peddle and Franklin 1991, Franklin and McDermid 1993, Ryherd and Woodcock 1996, Wulder et al 1998, Franklin et al 2001.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A multiscale texture analysis procedure for improved forest stand classification

Coburn

Roberts

2004

International Journal of Remote Sensing

229

135

View full text Add to dashboard Cite

Abstract. Image texture is a complex visual perception. With the everincreasing spatial resolution of remotely sensed data, the role of image texture in image classification has increased. Current approaches to image texture analysis rely on a single band of spatial information to characterize texture. This paper presents a multiscale approach to image texture where first and secondorder statistical measures were derived from different sizes of processing windows and were used as additional information in a supervised classification. By using several bands of textural information processed with different window sizes (from 565 to 15615) the main forest stands in the image were improved up to a maximum of 40%. A geostatistical analysis indicated that there was no single window size that would adequately characterize the range of textural conditions present in this image. A number of different statistical texture measures were compared for this image. While all of the different texture measures provided a degree of improvement (from 4 to 13% overall), the multiscale approach achieved a higher degree of classification accuracy regardless of which statistical procedure was used. When compared with single band texture measures, the level of overall improvement varied between 4 and 8%. The results indicate that this multiscale approach is an improvement over the current single band approach to analysing image texture.

show abstract

“…In a similar manner, they have also been used as a tool to model the spatial variation within images (Woodcock et al 1988b, St-Onge and Cavayas 1995, Wulder et al 1998, Treitz 2001.…”

Section: Geostatisticsmentioning

confidence: 99%

Section: Geostatisticsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 96%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A multiscale texture analysis procedure for improved forest stand classification

Coburn

Roberts

2004

International Journal of Remote Sensing

229

135

View full text Add to dashboard Cite

show abstract

“…Forest structure assessment using remote sensing is best conducted with both spectral and spatial image information, although the latter has not received as much attention in the literature as the former. Image spatial measures such as texture (e.g., Wulder et al, 1998;Olthof and King, 2000), semivariance parameters (e.g., Bowers et al, 1994;Le´vesque and King, 1999;Sampson et al, 2001), and image fractions (e.g., Peddle et al, 1999;Le´vesque and King, 2003) have been found to be useful for forest structure modelling and, in some cases, forest damage. Most of these studies, though, have been concerned with prediction of single forest variables.…”

Section: Sensingmentioning

confidence: 99%

Modelling and Mapping Damage to Forests from an Ice Storm Using Remote Sensing and Environmental Data

et al. 2005

View full text Add to dashboard Cite

Abstract. An extreme ice storm in January 1998 deposited up to 100 mm of ice and resulted in significant forest damage across eastern North America. Average crown loss of over 75% was recorded in large areas of eastern Ontario and southern Quebec. A primary question that arose following the storm was: can forest damage be effectively assessed using remote sensing and other available environmental data? This paper presents two contrasting studies to address this question. The first involves damage modelling at a local scale in an unmanaged forest using spectral and spatial information in high-resolution airborne imagery. Results of field data analyses are also given that show relations between damage and forest structure and composition as well as changes in forest structure that occurred in the years following the storm. The second study involves regional scale damage mapping in managed and unmanaged forests of eastern Ontario using medium resolution satellite imagery and other environmental data. In comparison of several image classification and data interpolation methods, the best damage map was produced using a neural network classifier and a mix of Landsat and environmental data. The methods and results presented in this paper form the basis for ongoing long-term temporal study of damage impacts on forest condition.

show abstract

Relationships between meridional profiles of satellite-derived vegetation index (NDVI) and climate over Siberia

2000

View full text Add to dashboard Cite

This study investigates the regional relationship between the satellite‐derived vegetation index (Normalized Difference Vegetation Index, NDVI) and climatological parameters (temperature and precipitation) over Siberia on a 5‐year (1986–1990) annual mean basis. The NDVI in Siberia shows a large value around the 60°N zone, and it gradually decreases toward the southern arid region and the low temperature polar region. This meridional profile (south–north regionality) of the NDVI was analysed in two meridional transects, along 75°E (arid–forest transect) and 110°E (forest–tundra transect). A modified warmth index (WI(0)) was utilized as a temperature index. In the 75°E transect, high positive (0.79) and negative (−0.58) correlations were found in the NDVI–precipitation and the NDVI–WI(0) meridional profiles, respectively. This fact implies that precipitation plays a role in providing water and dominates the vegetation distribution, while high temperature induces dryness and water shortages, i.e. the critical factor for the vegetation meridional profile is available water in the arid–forest transect. In the 110°E transect, a high positive (0.92) correlation was found, which suggests that the dominant factor for the NDVI profile is temperature in the forest–tundra transect, i.e. the critical factor for the vegetation meridional profile here may be temperature. Furthermore, these meridional profiles were scrutinized in terms of the station altitude and it was suggested that the NDVI tends to change depending on the WI(0) and precipitation at regional variations, which are basically the result of the altitude difference between the stations. The comprehensive relationship between the NDVI, WI(0) and precipitation regionalities is discussed in terms of six vegetation types. The result shows that high NDVI (>0.2) is observed when the WI(0) is over 40°C and precipitation is more than 2.5×WI(0)+50. Copyright © 2000 Royal Meteorological Society

show abstract

Aerial Image Texture Information in the Estimation of Northern Deciduous and Mixed Wood Forest Leaf Area Index (LAI)

Cited by 184 publications

References 48 publications

A multiscale texture analysis procedure for improved forest stand classification

A multiscale texture analysis procedure for improved forest stand classification

Modelling and Mapping Damage to Forests from an Ice Storm Using Remote Sensing and Environmental Data

Relationships between meridional profiles of satellite-derived vegetation index (NDVI) and climate over Siberia

Contact Info

Product

Resources

About