How it is working


PhotoModeler and photogrammetry are new to many of our website visitors, and some of you will not be familiar with some of the terms that appear within these pages The new PhotoModeler Web Glossary provides definitions of terms used in describing PhotoModeler and its applications.

The Links page provides some useful external resources and links.

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | Y | Z

Computer Aided Design, or Computer Aided Drafting.
Camera Calibration
Camera calibration is the process of finding the true parameters of the camera that took your photographs. Some of these parameters are focal length, format size, principal point, and lens distortion.
Camera Orientation
a) the position and angles of a camera relative to its reference coordinate frame, or b) the process whereby the program determines the position and angles of one or more Camera Stations.
Camera Station
A Camera Station is a single position of a camera when it took a photograph. Imagine freezing time just as the shutter was clicked to take a photograph - that frozen position and orientation of the camera is a “camera station”.
Coded Target
A target that has a unique code ring around it that PhotoModeler can automatically recognize. Coded Targets provide the ability to automatically mark, recognize and reference targets in a scene.
A Constraint is a piece of additional information the user knows (or is willing to assume) about the scene or object. For example, knowing that 5 points in a project all line in a straight line (i.e. are collinear) is a constraint. There are several types of Constraints in PhotoModeler: Multi-photo Constraints, Axes Constraints, and Control Point Constraints.
Control Points
A Control Point is a known location in space with either X,Y or X,Y,Z values known. Control Points can be used in photogrammetry to define a known coordinate system, or to form constraints on the solutions.
The DXF file format is a data exchange standard defined by AutoDesk, Inc. Many CAD and rendering packages can import DXF data files. PhotoModeler can create a 3D or 2D DXF file containing the solved 3D model. PhotoModeler can import Control Points from a DXF file.
^ Back to top
Short for EXchangeable Image File. This is a file format specification for storing information in digital photography image files. Almost all new digital cameras use the EXIF format, storing information in the image such as shutter speed, F number, focal length, date/time, resolution, etc. JPEG EXIF files can be read by any application supporting JPEG. Not all programs read the EXIF header information though (as PhotoModeler does).
A distinguishing mark on an object or in a scene that can be seen in a photo. We talk about marking and measuring features. Whether something is a feature or not is in the eye of the beholder. The edge of a window frame would be a feature to someone modeling window positions on the facade of a building.
In context of a camera in photogrammetry, a fiducial is a reference point that is seen in the resulting image that came from a point fixed to the camera body. Fiducials allow PhotoModeler to recover the internal reference frame of a film camera for maximum precision.
Focal Length
Focal Length is the Principal Distance when the lens of the camera is focused at infinity. Also see Principal Distance.
Format Aspect Ratio
The Format Aspect Ratio is defined as the camera's format width divided by its format height.
Format Size
The size of a camera's imaging area in millimeters. In a digital camera, it is the size of the area of the imaging chip (known as the CCD). In a 35mm film camera has a format size of approximately 36x24mm.
Full Frame Prints
Full frame scans or prints show the whole 35mm frame (the 36mm x 24mm black frame around the picture). If the scan or print shows the sprocket holes then you have full frame prints. Most 35mm prints are not full-frame prints. Full frame prints can be used more accurately in PhotoModeler if there are no other fiducials.
^ Back to top
Inverse Camera
The process of computing a cameras parameters from the scene using control points. This is a form of calibration but we do not call it 'calibration' as Inverse Camera requires constraints (usually Control Points), tends not to be as accurate, and is used only in cases where one or more photographs are being used from an unknown camera (e.g. from a bystander or achieves).
Lens Distortion
All camera lenses distort images to some extent. “Barrel distortion” is a type of lens distortion that is commonly seen. In the extreme this would make a cylindrical barrel look like an old wooden barrel (with curved sides). Straight lines (esp. at the edges of the image frame) become curved. If lens distortion is not taken into account by a photogrammetric program the resulting accuracy is greatly reduced. PhotoModeler does compensate for both radial and de-centering lens distortion to give the highest accuracy measurements and models.
A location on a single photograph representing the image of some physical feature.
Marking Residuals
A residual in photogrammetry is short for "residual error" - that is, the error left over in some value after processing has completed. A "marking residual" is the difference between where PhotoModeler thinks a point should be on the photograph and where you have marked it. This is calculated by projecting the 3D position of the point onto the photograph and comparing it with the marked position. The smaller the marking residual usually the more precise the point has been marked.
A 2D or 3D curve in space. NURBS = Non-Uniform Rational B-Spline. NURBS are industry standard tools for the representation and design of geometry.
NURBS Surface
3D surface with a smooth curved shape. NURBS = Non-Uniform Rational B-Spline. NURBS are industry standard tools for the representation and design of geometry.
^ Back to top
Object Point
A 3D Point that is created by referencing two or more Marked Points across photographs. 3D Point and Object Point are used interchangeably.
Orient a Photo / Camera Station
Orientation is the process of finding the location and angles of a camera at the time it took the photo.
Orthophoto or Orthographic Photo
An orthophoto (sometimes ortho-photo) is an image in which all perspective related distortions have been removed. It would be similar to taking a photograph of a surface where the camera is perpendicular to the surface and an infinite distance away. Orthophotos are like orthographic drawings in that they can be used to do distance and angle measurements. Every object that is the same size in the real world will be the same size in the orthophoto regardless of distance. Compare to a standard photograph with perspective distortion where as objects get farther from the camera they get smaller. PhotoModeler can create orthophotos.
Perspective Matching
The process of determining the position of a camera when it took a photograph so that 3D graphics can be combined with that photograph. To properly combine 3D graphics with a photograph you have to tell the 3D rendering program where to place the camera, how to orient the camera, and the specification for the camera's internal parameters so during rendering the 'perspective matches'.
Photo Texture
A mapping of a photograph onto a surface during 3D rendering to increase the realism of that rendering.
The art and science of obtaining measurements from photographs. PhotoModeler is a photogrammetry software program.
Principal Distance
The distance between the imaging surface plane (piece of film in a 35mm camera or CCD chip in a digital camera) and a point where all light rays intersect inside the lens. A principal distance of 50mm means that the distance from the point where all the light rays focus to the imaging plane is 50mm long. PhotoModeler uses Focal Length to mean Principal Distance for ease of understanding.
Principal Point
The location in a camera where the optical axis of the lens intersects the imaging plane. It is the reference point in the image to which all marks and lens distortion parameters are related.
Either the process or the end result of displaying the rendered 3D model superimposed on the project’s photographs. That is the 3D model is rendered from exactly the same view point as the photograph so they match up.
^ Back to top
RAD Targets
RAD Targets are the new style of coded target introduced in 2009. RAD stands for Ringed Automatically Detected. These new coded targets are more robust and are detected more quickly. See Coded Target.
Rectified Photograph
A rectified photograph is a remapping of a single plane viewed in perspective to an orthographic view. Rectified Photographs are used to turn a slanted photograph of a building facade into a straight-on view suitable for doing drawings and measurements. A rectified photograph is an Orthophoto if the object plane is perfectly flat.
The process by which you identify to PhotoModeler all the marks, on all the photographs, that refer to the same physical point on the object you are measuring. In some fields this is called “feature matching”.
Retro-Reflective Targets
A type of target made from retro-reflective material. Retro-reflective material reflects light back in the direction of the light source. This is useful in photogrammetry because if you place a strong light source at the camera (such as a flash) then regardless of the angle of the target, the target will strongly reflect this light. Retro-reflective targets are easy to identify, can be precisely marked, and are suited to working at night or in other darkened environments.
Self-Calibration / Field Calibration
The process of fine tuning an existing camera calibration during the processing of a standard project. With Self-Calibration PhotoModeler makes minor adjustments to the camera parameters on a per-photograph basis to account for changes in focal length, principal point and lens distortion as the camera lens is focused. With Field Calibration PhotoModeler makes adjustments to the camera parameters assuming the camera was consistent across all photos. Field Calibration and Camera Calibration use the same adjustment algorithm but in the Field Calibration case it is being done at the same time as a real project is being solved. For high accuracy projects, Field Calibration (along with appropriate photographs) can be very important in achieving that extra bit of accuracy.
SmartMatch is a new tool introduced with the PhotoModeler Scanner 2010 version. It automatically detects feature points in images and matches them. It does complete project setup (after importing photos, you press a button and the result is a fully oriented set a photos) without coded targets. It is like automation for normal images without targets.
Sub-Pixel Target Marking
Sub-Pixel Target Marking is a feature that allows you to mark points in an accurate and consistent manner with the use of targets. By knowing a priori what the target is going to look like, the software can study the digital image data and mark the target precisely.
Surface Draw
Surface Draw provides the ability to create 3D data by marking in only one photograph. This is done by tracing details on an area of a photograph that is covered by a projected 3D surface. No referencing is required. Surface Draw is useful for modeling detailed patterns that lie on one plane, such as brick work on the facade of a building. Surface Drawing is also useful for drawing curves that lie on planes from a single photograph, such as a skid mark on a road surface.
A target is a point in a scene or on an object that can be easily identified and easily marked when viewed in an image. Typically, a target is a high contrast dot or filled circle. Such a target can be marked with high precision with the sub-pixel marker (manual or automatic).


© 2012 Eos Systems Inc.