Cameras table schema

This is a description of the cameras table of the frame camera raster type with a list of supported fields. A frame camera refers to a single perspective imaging camera with an optical lens that can be modeled as a pinhole. The frame camera raster type allows you to manage frame camera imagery in a mosaic dataset.

Cameras table

To use frame camera imagery in a mosaic dataset, you must provide detailed camera parameters and frame information that define a mathematical relationship between the image coordinate system, camera coordinate system, and ground coordinate system.

The following diagram shows how pixels are transformed between the coordinate systems:

Transformations for frame camera imagery

The cameras table contains parameters that are unique to each camera, such as focal length, principal point of autocollimation, X and Y coordinates, and affine transformation coefficients. These parameters define the interior orientation of the camera. The table format can be a geodatabase table, a feature class table, or a .csv file.

You can also use the Build Frames & Cameras Tables tool to create the cameras table in a guided process.

Cameras table fields

The supported cameras table fields are listed below. Note that a parameter definition that is duplicated in the frames table will take precedence over the definition in the cameras table.

Fields supported in the cameras table

Field name Field typeData typeDescriptionDomain

ObjectID

Required

Numeric

The unique number identifier for each camera.

CameraID

Required

String

The primary key identifying the camera parameters.

FocalLength

Required

Numeric

The focal length of the camera lens, measured in microns.

PrincipalX

Optional

Numeric

The x-coordinate of the principal point of the autocollimation, measured in microns. If not defined, the default is 0.

The principal point is the offset between the fiducial center and the principal point of autocollimation (PPA). The principal point of symmetry (PPS) is assumed to be the same as the PPA.

PrincipalY

Optional

Numeric

The y-coordinate of the principal point of the autocollimation, measured in microns. If not defined, the default is 0.

The principal point is the offset between the fiducial center and the PPA. The PPS is assumed to be the same as the PPA.

BlockName

Optional

String

The name of the block (project) where the image belongs.

NRows

Optional

Numeric

The number of rows of pixels in the image.

NColumns

Optional

Numeric

The number of columns of pixels in the image.

NBands

Optional

Numeric

The number of bands of pixels in the image.

PixelType

Optional

Numeric or String

The pixel type for the image, either as a numeric value matching an rstPixelType or a matching string.

Numeric value matching an rstPixelType: PT_U1=0, PT_U2=1, PT_U4=2, PT_UCHAR=3, PT_CHAR=4, PT_USHORT=5, PT_SHORT=6, PT_ULONG=7, PT_LONG=8, PT_FLOAT=9, PT_DOUBLE=10, PT_COMPLEX=11, PT_DCOMPLEX=12, PT_CSHORT=12, PT_CLONG=14.

Matching string: 8_BIT_UNSIGNED, 8_BIT_SIGNED, 16_BIT_UNSIGNED, 16_BIT_SIGNED, 32_BIT_UNSIGNED, 32_BIT_SIGNED, 32_BIT_FLOAT, 1_BIT, 2_BIT, 4_BIT, 64_BIT.

PixelSize

Optional

Numeric

The sensor pixel size.

Unit is micron.

FilmCoordinateSystem (FCS)

Optional

Integer

Defines the film coordinate system of a scanned aerial photograph, and digital aerial camera. It is used in computing fiducial information and affine transformation construction.

Note:

Common practice of the digital aerial camera case is the camera's X is aligned with flight direction (option 1, X_RIGHT_Y_UP, the default). FCS is used to accommodate non standard cases.

  • 1 — X_RIGHT_Y_UP, where the origin of the scanned photo's coordinate system is the center, and positive X points right and positive Y points up. This is the default.
  • 2 — X_UP_Y_LEFT, where the origin of the scanned photo's coordinate system is the center, and positive X points up and positive Y points left.
  • 3 — X_LEFT_Y_DOWN, where the origin of the scanned photo's coordinate system is the center, and positive X points left and positive Y points down.
  • 4 — X_DOWN_Y_RIGHT, where the origin of the scanned photo's coordinate system is the center, and positive X points down and positive Y points right.

SRS

Optional

String

The coordinate system associated with the perspective point as a file path or WKID (EPSG code). For EPSG code, the coordinate system for x,y and z are separated by a semicolon (;), for example 26918;5773. If not defined, it defaults to the user-specified coordinate system or the coordinate system defined in the mosaic data spatial reference. The spatial reference can be defined for each image item in the mosaic dataset as well. If the SRS parameter is defined in both the cameras table and the frames table, the value in the frames table will take precedence.

OrientationType

Optional

String

Specifies how the rotational Exterior Orientation (EO) parameters are described. The default is OPK.

OPK—Indicates that the rotational EO parameters are defined as angles in the Omega, Phi, Kappa, Angle Direction, and Polarity fields.

Matrix—Indicates that the rotational EO parameters are defined as a matrix of nine coefficients in the Matrix field.

AverageZ

Optional

Numeric

The average ground height. The default is the value specified in the raster type's orthorectification properties, or zero if it's not specified.

ApplyECC

Optional

Boolean

Indicates whether to account for the curvature of the earth when applying transformations with respect to ground coordinates. The default is FALSE.

True—Accounts for the curvature of the earth when applying the transformations with respect to ground coordinates.

False—Assumes the earth is flat.

EarthRadius

Optional

Numeric

An alternative value to use for the curvature of the earth adjustment. The default value is 6378137.0 meters.

Units are meters.

AngleDirection

Optional

String

Specifies the direction of the EO angles. The default value is -1.

-1: Indicates that the EO angles are specified in the clockwise direction.

+1: Indicates that the EO angles are specified in the counterclockwise direction.

Polarity

Optional

Numeric

Indicates whether the image plane is presumed to be on the same or the opposite side of the perspective center as the object or ground plane. The default is-1, which is the opposite side.

-1: Indicates the opposite side of the image-plane.

+1: Indicates the same side of the image-plane.

DistortionType

Optional

String

Specifies how the lens distortion is described. The default is DistortionModel.

The default is Distortion Model.

Distortion Model—Indicates that the distortion correction is described by the coefficients defined in the Radial and Tangential fields.

Distortion Table—Indicates that distortion is defined as a set of (r,v) pairs indicating radial distance and corresponding distortion value. The RadialDistances and RadialDistortions. fields must be populated.

Radial

For Distortion Type = DistortionModel

String

Specifies the set of four space- or semicolon-delimited coefficients describing radial distortion, for example, 0;0;0;0 for K0;K1;K2;K3.

The coupling unit is millimeter (mm).

If there are three radial distortion parameters provided, leave k0 equal to zero, and enter values for k1, k2, and k3.

If parameters are provided with no coupling unit, which is common in computer vision applications, convert the parameters using the following equations, where f is the focal length in millimeters and K1_cv is the computer vision parameter without a coupling unit:

  • K1_mm = K1_cv / (f * f)
  • K2_mm = K2_cv / (f * f * f * f)
  • K3_mm = K3_cv / (f * f * f * f * f * f)

Tangential

For Distortion Type = DistortionModel

String

Specifies the set of two space- or semicolon-delimited tangential distortion coefficients, for example, 0;0 for P1;P2.

The coupling unit is millimeter (mm).

If parameters are provided with no coupling unit, convert the parameters using the following equations, where f is the focal length in millimeters and p1_cv is the computer vision parameter without a coupling unit:

  • P1_mm = -p1_cv/f
  • P2_mm = -p2_cv/f

RadialDistances

Optional

String

The space- or semicolon-delimited radial distances are indicated as an ordered set of N values <r[i]>. Each r[i] distance value has a corresponding d[i] distortion value in the RadialDistortions field.

The unit is micrometers.

RadialDistortions

Optional

String

The space- or semicolon-delimited distortion values are indicated as an ordered set of N values <d[i]>. Each d[i] distortion value corresponds to the distance value r[i] in the RadialDistance field.

The unit is micrometers.

FilmFiducials

Optional

Numeric

Stores fiducial coordinates for the camera in microns.

The format is value pairs separated by semicolons, for example, "106003.0 -106000.0;-105997.0 -106001.0;-106002.0 105998.0;105999.0 106000.0".

This example shows four corner fiducials, whereas some cameras may have four edge fiducials, while other cameras may have 8 fiducial locations.

AffineDirection

Optional

String

Specifies the direction of the IO affine transformation. If not specified, a default direction of image-to-film (+1) is assumed.

+1: image-to-film.

-1: film-to-image.

A0, A1, A2 and B0, B1, B2 fields

Optional. Not required when pixel size and FCS are defined.

Numeric

The coefficient of the affine transformation that establishes a relationship between image space and film space. The direction of this transformation is indicated using the AffineDirection field. If that field does not exist, the default is image space to film space. The units for this are microns.

A0, A1, A2 represents the translation in x direction.

B0, B1, B2 represents the translation in y direction.

Note:

Instead of calculating A0, A1, A2 and B0, B1, and B2, it is recommended that you define the pixel size of the camera, which is used to automatically determine the affine parameters. The camera's pixel size is typically supplied with a camera's calibration information. If the camera's pixel size is not provided, you can compute it using the equation below:

Pixel Size = CCD Diagonal / Image Diagonal(in pixels)

where

CCD Diagonal = 2 * (Focal Length * Tan(FOV/2))
Note:

The equations for computing the corrections for radial distortion and tangential distortion.

Cameras table example

A sample cameras table stored as a geodatabase table is shown below. This table references the possible fields above. The sample contains one row, since only one camera was used.

Sample cameras table

OBJECTIDCameraIDFocal Length (µm)Principal X (µm)Principal Y (µm)Pixel Size (µm)

1

UltraCamXp_Pan

100500

-120

0

6

2

UltraCamXp_MS

100500

0

0

18

Related topics