Name: BuildingFootprints

Display Field: Id

Type: Feature Layer

Geometry Type: esriGeometryPolygon

Description: The LAS data set was originally classified according to 4 classes (ground, water, bridge overpass, and noise), with the rest of the data being unclassified. That left some classes to be derived and classified, of which one—the building/ structure class—was considered necessary for this project. In theory, deriving a building/structure layer is relatively straightforward: the building reflectance response should be unclassified, single-reflectance response points, whereas the vegetation, also unclassified, should yield a multiple-reflectance response as the beam bounces back through the canopy. Following this idea, we created a Digital Surface Model (DSM) from the single-response, unclassified LAS point cloud. We then subtracted these DSMs from the Bare Earth DEMs to create a difference image, which ideally should represent only buildings. Unfortunately, many trees were included in this “buildings” layer, due possibly to the sparse canopy that is characteristic of trees found in southwestern forests and possibly to the presence of fairly recent burn scars that include a number of standing dead trees and snags. In an attempt to remove the clutter of false positives due to trees, we developed a Normalized Difference Vegetation Index (NDVI) from the NAIP imagery acquired over the area in the same year. The NDVI is an image-processing technique that uses the reflective information found in the red (Red) and near-infrared (NIR) wavelengths to enhance the “green” vegetative response over other, non-vegetated surface features (Eq. 1). NDVI = (NIR−Red)/(NIR+Red) [Eq. 1]. This provides a floating-point image of values from -1 to 1, with numbers above 0 representing increasing vegetative cover. We further modified the NDVI equation to create an 8-bit image (Eq. 2). NDVImod = (NDVI+1)*100 [Eq. 2]. This 8-bit image had all positive integer values, where values above 100 indicated increasing vegetative cover. We used the generated NDVI image, in particular values above 109, to mask out many of the false anomalies. In addition, all heights less than 6 feet were masked out, as this was considered a minimum height for most buildings. We added 1 to values in the resulting image so that all values, even the zeroes, would be counted. Then values were clumped to produce an image of individually coded raster polygons. We eliminated all clusters smaller than 32 square meters (345 square feet) from the clumped image, ran a 3x3 majority filter to remove relict edges, and ran a 3x3 morphological close filter to remove holes in the raster polygons. We completed the raster processing in ERDAS IMAGINE and then converted the data set to a polygon layer in ESRI ArcGIS, as is and without using the ‘simplify polygon’ option. This was cleaned up further using the simplify buildings module with a minimum spacing of 2 meters. Once this was completed, the polygon layer was edited using the NAIP imagery and DSM Shaded Relief imagery as a background by a heads-up digitizing at a 1:3,000 scale (the approximate base resolution of the LiDAR data). The building/structure layer contained more than 44,612 identified structures.

Definition Expression: N/A

Copyright Text: Earth Data Analysis Center

Default Visibility: true

MaxRecordCount: 1000

Supported Query Formats: JSON, AMF, geoJSON

Min Scale: 5000

Max Scale: 0

Supports Advanced Queries: true

Supports Statistics: true

Has Labels: false

Can Modify Layer: true

Can Scale Symbols: false

Use Standardized Queries: true

Supports Datum Transformation: true

Extent:

XMin: 654519.2217508294
YMin: 943133.9232339859
XMax: 934452.5181823298
YMax: 1441207.1249202415
Spatial Reference: 102712  (2257)

Drawing Info:

Renderer:
Simple Renderer:
Symbol:
Style: esriSFSSolid
Color: [104, 104, 104, 255]
Outline:
Style: esriSLSSolid
Color: [110, 110, 110, 255]
Width: 0
Label:
Description:
Transparency: 0
Labeling Info:

Advanced Query Capabilities:

Supports Statistics: true
Supports OrderBy: true
Supports Distinct: true
Supports Pagination: true
Supports TrueCurve: true
Supports Returning Query Extent: true
Supports Query With Distance: true
Supports Sql Expression: true

HasZ: false

HasM: false

Has Attachments: false

HTML Popup Type: esriServerHTMLPopupTypeAsHTMLText

Type ID Field: null

Fields:

• OBJECTID ( type: esriFieldTypeOID , alias: OBJECTID )
• Id ( type: esriFieldTypeInteger , alias: Id )
• Shape_Leng ( type: esriFieldTypeDouble , alias: Shape_Leng )
• InBld_FID ( type: esriFieldTypeInteger , alias: InBld_FID )
• Shape ( type: esriFieldTypeGeometry , alias: Shape )
• Shape.STArea() ( type: esriFieldTypeDouble , alias: Shape.STArea() )
• Shape.STLength() ( type: esriFieldTypeDouble , alias: Shape.STLength() )

Supported Operations:   Query   Generate Renderer   Return Updates