John R. Jensen

Object-based change detection using correlation image analysis and image segmentation

This study introduces change detection based on object/neighbourhood correlation image analysis and image segmentation techniques. The correlation image analysis is based on the fact that pairs of brightness values from the same geographic area (e.g. an object) between bi‐temporal image datasets tend to be highly correlated when little change occurres, and uncorrelated when change occurs. Five different change detection methods were investigated to determine how new contextual features could improve change classification results, and if an object‐based approach could improve change classification when compared with per‐pixel analysis. The five methods examined include (1) object‐based change classification incorporating object correlation images (OCIs), (2) object‐based change classification incorporating neighbourhood correlation images (NCIs), (3) object‐based change classification without contextual features, (4) per‐pixel change classification incorporating NCIs, and (5) traditional per‐pixel change classification using only bi‐temporal image data. Two different classification algorithms (i.e. a machine‐learning decision tree and nearest‐neighbour) were also investigated. Comparison between the OCI and the NCI variables was evaluated. Object‐based change classifications incorporating the OCIs or the NCIs produced more accurate change detection classes (Kappa approximated 90%) than other change detection results (Kappa ranged from 80 to 85%).

An evaluation of LIDAR- and IFSAR-derived digital elevation models in leaf-on conditions with USGS Level 1 and Level 2 DEMs

Abstract An assessment of four different remote sensing based methods for deriving digital elevation models (DEMs) was conducted in a flood-prone watershed in North Carolina. New airborne LIDAR (light detecting and ranging) and IFSAR (interferometric synthetic aperture radar (SAR)) data were collected and corresponding DEMs created. These new sources were compared to two methods: Gestalt Photomapper (GPM) and contour-to-grid, used by the U.S. Geological Survey (USGS) for creating DEMs. Survey-grade points (1470) for five different land cover classes were used as reference points. One unique aspect of this study was the LIDAR and IFSAR data were collected during leaf-on conditions. Analyses of absolute elevation accuracy and terrain slope were conducted. The LIDAR- and contour-to-grid derived DEMs exhibited the highest overall absolute elevation accuracies. Elevation accuracy was found to vary with land cover categories. Elevation accuracy also decreased with increasing slopes—but only for the scrub/shrub land cover category. Appreciable terrain slope errors for the reference points were found with all methods.

Synergistic use of lidar and color aerial photography for mapping urban parcel imperviousness

The imperviousness of land parcels was mapped and evaluated using high spatial resolution digitized color orthophotography and surface-cover height extracted from multiple-return lidar data. Maximum-likelihood classification, spectral clustering, and expert system approaches were used to extract the impervious information from the datasets. Classified pixels (or segments) were aggregated to parcels. The classification model based on the use of both the orthophotography and lidar-derived surface-cover height yielded impervious surface results for all parcels that were within 15 percent of reference data. The standard error for the rule-based per-pixel model was 7.15 percent with a maximum observed error of 18.94 percent. The maximum-likelihood per-pixel classification yielded a lower standard error of 6.62 percent with a maximum of 14.16 percent. The regression slope (i.e., 0.955) for the maximum-likelihood per-pixel model indicated a near perfect relationship between observed and predicted imperviousness. The additional effort of using a per-segment approach with a rule-based classification resulted in slightly better standard error (5.85 percent) and a near-perfect regression slope (1.016).

Effectiveness of Subpixel Analysis in Detecting and Quantifying Urban Imperviousness from Landsat Thematic Mapper Imagery

Abstract The quantification of urban imperviousness using remotely sensed spectral data is next to impossible unless a spectral fraction of impervious components in an urban pixel can be detected. In this research, a multiple‐signature subpixel analysis technique coupled with a layered classification approach was developed to map urban imperviousness of each pixel of an urban scene into eight 10‐percent levels. The subpixel analysis was based on the idea of removing background spectra from the total radiance of a pixel and testing the residual spectrum against the signature spectrum. This study demonstrated that although the subpixel analysis was able to quantify urban imperviousness from most of the urban pixels, it experienced some difficulty in handling the spectral heterogeneity of diverse urban features. The layered classification approach was used to identify the extreme cases. An experiment of classifying Landsat TM data into eight levels of urban imperviousness revealed that 83.0% (kappa = 0.787) ...

An Evaluation of Lidar-derived Elevation and Terrain Slope in Leaf-off Conditions

The effects of land cover and surface slope on lidar-derived elevation data were examined for a watershed in the piedmont of North Carolina. Lidar data were collected over the study area in a winter (leaf-off) overflight. Survey-grade elevation points (1,225) for six different land cover classes were used as reference points. Root mean squared error (RMSE) for land cover classes ranged from 14.5 cm to 36.1 cm. Land cover with taller canopy vegetation exhibited the largest errors. The largest mean error (36.1 cm RMSE) was in the scrub-shrub cover class. Over the small slope range (0° to 10°) in this study area, there was little evidence for an increase in elevation error with increased slopes. However, for low grass land cover, elevation errors do increase in a consistent manner with increasing slope. Slope errors increased with increasing surface slope, under-predicting true slope on surface slopes � 2°. On average, the lidarderived elevation under-predicted true elevation regardless of land cover category. The under-prediction was significant, and ranged up to � 23.6 cm under pine land cover.

Predictive modelling of coniferous forest age using statistical and artificial neural network approaches applied to remote sensor data

Age is a powerful variable that can be of significant value when modelling the health of forest-dominated ecosystem. Traditional investigations have attempted to extract age information from remotely sensed data by regressing the spectral values with in situ derived age data. Traditional statistical approaches assume (a) normally distributed remote sensing and in situ data, (b) no collinearity among variables, and (c) linear data relationships. Artificial neural networks (ANNs) are not bound by such assumptions and may yield improved predictive modelling of forest stand biophysical parameters if properly utilized. This study investigated traditional statistical and ANN approaches to perform the predictive modelling of the age of loblolly pine (Pinus taeda) for large stands in southern Brazil using Thematic Mapper (TM) data. An extensive comparison of pattern associator and back-propagation ANNs versus both linear and nonlinear regression analysis was conducted. Various neural network architectures were investigated to determine the optimal configuration for this particular dataset. Certain back-propagation ANNs modelled stand age significantly better than traditional statistical approaches because of their ability to take into account nonlinear, non-normally distributed data. The results suggest that ANN analysis may be of significant value when using remote sensing data to model certain forest variables.

Integrating LIDAR elevation data, multi-spectral imagery and neural network modelling for marsh characterization

Vertical elevation relative to mean sea level is a critical variable for the productivity and stability of salt marshes. This research classified a high spatial resolution Airborne Data Acquisition and Registration (ADAR) digital camera image of a salt marsh landscape at North Inlet, South Carolina, USA using an artificial neural network. The remote sensing‐derived thematic map was cross‐referenced with Light Detection and Ranging (LIDAR) elevation data to compute the frequency distribution of marsh elevation relative to tidal elevations. At North Inlet, the median elevation of the salt marsh dominated by Spartina alterniflora was 0.349 m relative to the North American Vertical Datum 1988 (NAVD88), while the mean high water level was 0.618 m (2001 to May, 2003) with a mean tidal range of 1.39 m. The distribution of elevations of Spartina habitat within its vertical range was normal, and 80% of the salt marsh was situated between a narrow range of 0.22 m and 0.481 m. Areas classified as Juncus marsh, dominated by Juncus roemerianus, had a broader, skewed distribution, with 80% of the distribution between 0.296 m and 0.981 m and a median elevation of 0.519 m. The Juncus marsh occurs within the intertidal region of brackish marshes and along the upper fringe of salt marshes. The relative elevation of the Spartina marsh at North Inlet is consistent with recent work that predicts a decrease in equilibrium elevation with an increasing rate of sea‐level rise and suggests that the marshes here have not kept up with an increase in the rate of sea‐level rise during the last two decades.

Impact of Lidar Nominal Post-spacing on DEM Accuracy and Flood Zone Delineation

Lidar data have become a major source of digital terrain information for use in many applications including hydraulic modeling and flood plane mapping. Based on established relationships between sampling intensity and error, nominal post-spacing likely contributes significantly to the error budget. Post-spacing is also a major cost factor during lidar data collection. This research presents methods for establishing a relationship between nominal post-spacing and its effects on hydraulic modeling for flood zone delineation. Lidar data collected at a low post-spacing (approximately 1 to 2 m) over a piedmont study area in North Carolina was systematically decimated to simulate datasets with sequentially higher post-spacing values. Using extensive first-order ground survey information, the accuracy of each DEM derived from these lidar datasets was assessed and reported. Hydraulic analyses were performed utilizing standard engineering practices and modeling software (HEC-RAS). All input variables were held constant in each model run except for the topographic information from the decimated lidar datasets. The results were compared to a hydraulic analysis performed on the un-decimated reference dataset. The sensitivity of the primary model outputs to the variation in nominal post-spacing is reported. The results indicate that base flood elevation does not statistically change over the post-spacing values tested. Conversely, flood zone boundary mapping was found to be sensitive to variations in post-spacing.

The measurement of mangrove characteristics in southwest Florida using spot multispectral data

Abstract An intensive in situ sampling program near Marco Island, Florida during 19–23 October 1988 collected information on mangrove type, maximum canopy height, and percent canopy closure. These data were correlated with selected vegetation index information derived from analysis of SPOT multispectral (XS) data obtained on 21 October 1988. The Normalized Difference (ND) vegetation index information was the most highly correlated index with percent canopy closure (r=0.91). Percent canopy closure information can be used as a surrogate for mangrove density which is of great value when predicting which parts of the mangrove ecosystem are at greatest risk after an oil spill occurs. Such information is very valuable when constructing oil spill Environmental Sensitivity Index (ESI) Maps for tropical regions of the world.

Application of remote sensing and geographic information systems to the delineation and analysis of riparian buffer zones

Abstract Non-point source pollution has a significant impact on the quality of water resources. Studies have revealed that agricultural activities are often major contributors to non-point source pollution of aquatic environments. A common means of reducing the threat of non-point source pollution is through the establishment of riparian vegetation strips (or buffers) along those areas of stream channels that would be most susceptible to the threat. Remote sensing and geographic information systems (GIS) offer a means by which ‘critical’ areas can be identified, so that subsequent action toward the establishment of riparian zones can be taken. This research focuses on the development and analysis of riparian buffer zones for a portion of the Iowa River basin. Landsat Thematic Mapper (TM) data were used to characterize the land cover for the study area. An updated hydrology data layer was developed by integrating the United States Geological Survey (USGS) Digital Line Graph (DLG) data base with the TM-derived classification of surface water bodies. Spatial distance search tools were applied to develop the buffer zones around all surface hydrologic features. The buffer zones were integrated with the remotely sensed classification data to identify ‘critical’ areas for the establishment of riparian vegetation strips. Results indicated that while most of the main channel of the Iowa River was protected by natural vegetation, more than 44% (or 1008 ha) of the area along its tributaries lack any protective cover from non-point source pollution. As these ‘critical’ areas are adjacent to agricultural fields it is important that water resources management strategies focus on the establishment of riparian zones in order to minimize the impact of non-point source pollution.