HAP - Your Questions Answered


  • Input imagery refers to the positive digital image file (after scanning). Both uncomressed TIFF and PCI PIX files are supported. Note: A standard naming convention should be applied. This helps to ensure that the correct metadata is associated with the correct input image. For example: RollNumber-ImageNumber.tif, ProjectID_ImageNumber.tif, etc.
  • Camera characteristics are similar (i.e. similar focal length, dimensions)
  • Same Fiducial mark locations (i.e. corner, edge, both)
  • Near continuous coverage

For the HAP system to operate properly, the following should be considered as a minimum amount of metadata in your file(s):

  • Approximate Scene Center Coordinates
  • Camera Focal Length
  • Physical Paper Dimensions
  • Approximate Flying Altitude

Reference Data can be any imagery used for GCP collection, road vector layers used for GCP collection and a Digital Elevation Model (DEM) used for GCP collection and orthorectification. Before choosing your reference data, it is important to ask yourself the following questions regarding its suitability:

  • Can an appropriate control image be obtained that covers the project area(s)?
  • Can road vector layers be obtained for the project area(s)?
  • Can an appropriate DEM be obtained for the project area(s)?

For roll-film, scanners will need a specific attachment to hold and advance the spools. This attachment is usually provided with the scanner or purchased as an accessory. For flat-bed scanners, ensure the scanner is large enough to hold your film or print.

  • The scanning bed dimensions should be sufficient to fit the valid image area and the surrounding boarder
  • The border-areas of image prints are important as they may contain fiducial marks or information about the sensor (e.g. focal lengths, scale, etc.)
  • Common aerial image print dimensions are 9”x9”, 10”x10” and sometimes larger. Similarly rolls will be 9” or 10” spools of varying length.

Scanning resolution or dots per inch (DPI) follows the general rule that the higher the DPI, the more detail your scanned images will retain, but the scanning time per print and file-size will increase. It is important to determine a sufficient DPI that ensures the required detail from the image print is obtained, but also takes into account scanning speed and file size.

  • Scanning with a DPI that is set too high (i.e. 1000dpi) may be counterproductive as the detail of the image prints do not require that level of scan resolution
  • From discussion with industry partners, 600dpi is a very common scan resolution used for historical aerial imagery. However, in some cases a higher resolution is required (i.e. 1000dpi).
  • Please note, that contrary to common belief the DPI you scan at will not change the ground resolution (GSD) of the image, but will impact the number of pixels in the digital image file and therefore, the file size itself.
  • It may not seem important at first to consider whether your scanned images will be 300mb or 800mb, but if you have a large archive of images, file storage may be a problem.

Speed is a production concern (how many frames do you expect to scan an hour, etc). Scanners specifications with indicate how many seconds it takes to scan a file of a given size and at what dpi. Please make sure to select a scanner or service that will allow you to meet your quota and product specification

Photo alignment on the scanner bed is important to consider. It is important that the image prints are placed as square as possible and that there is some level of consistency to the placement of the images when scanning. This helps with automating the fiducial mark collection during HAP processing.

  • It is not important that the images be placed perfectly square as fiducial mark collection ensures that the image boundaries are correctly defined when processing with the HAP system.




Social Media

Free Trials

  • geomatica-50px  Geomatica
  • gxl-50px  Learn More About GXL
  • 3DFE  3D FeatureXtract

Latest Tweets

RT @GISResources: @PCIGeomatics Marks Milestone in #Cloud-Based #ImageProcessing #geospatial #GeoNews https://t.co/3qToUiJUxN
RT @gischethans: .@pcigeomatics #CEO says software adoption model is changing from perpetual licensing to subscription model... #GWF2017