2. Review of Survey Techniques

[Back to TOC] [Previous] [Next]


2.1 Approach

[Back to TOC] [Previous] [Next]

Numerous interviews of individuals involved with aerial video surveys were conducted in January and February 1995. The list of summaries of the Interviews (Appendix A) and Contacts (Appendix B) are provided. Information from these interviews was used to categorize the survey techniques and identify common problems associated with the techniques.


2.2 Overview of Techniques

[Back to TOC] [Previous] [Next]

The basic approach of aerial video imaging surveys is to use an video camera to record land imagery from an aerial platform. The technique has gained popularity over the past ten years because of the ability of users to tailor the imagery to their particular application. In comparison to conventional aerial photography, (a) video imagery can be acquired from a wide variety of aircraft and is usually not ceiling limited (i.e., AVI flying altitude is usually much lower than that required for conventional aerial photography), (b) synchronous narrative descriptions of resources can be added, greatly enhancing the interpretative quality of the imagery and (c) the oblique imagery is similar to the view from an aircraft and easily interpreted by the lay-person. The major disadvantage in comparison to conventional aerial photography is that AVI can generally not be used for planimetric mapping without special attention to aircraft attitude during the survey.

The ease of acquisition has inspired a wide range of applications to resource mapping within BC. The technique is most suitable to mapping linear features such as rivers, coastlines or pipelines that can be centered in the image. Examples of AVI application include: coastal mapping for oil spill sensitivity assessments, inventorying stream morphology for assessing fish habitat, inventorying shellfish habitats, coastal landscape cut-block planning and general documentation of resource conditions. The coastal mapping programs have been one of the most extensive, with an estimated 25% of the BC coastline imaged. Stream classification inventories have accounted for more than 20 surveys within the Province.

Other concurrent report compilations provide a review of aerial surveillance techniques for use in watershed management (see Ham 1995).

Reconnaissance AVI Surveys

A review of the existing surveys indicates three general classes of surveys have been conducted (Table 1). The authors' reviewed hundreds of hours of overview or Reconnaissance AVI as part of litigation support for the Exxon Valdez oil spill, and although little of this type of imagery has been formally reported for BC, such imagery is routinely collected by resource managers and news agencies. The greatest weakness of this type of imagery is the lack of location information; the user is left to estimate the original flightline, based on features evident in the image. Given the general uniformity of landforms in BC (i.e., conifer forests with few distinct cultural features), this lack of position information is a severe limitation to a broader-based use of the imagery.

Inventory AVI Surveys

Inventory AVI Surveys are the most common type of aerial video survey. These are surveys where imagery is acquired of a feature or set of features for a specific application. The survey is usually pre-planned and flightline information is known such that the features can be interpreted and approximately mapped (referred to as reconnaissance mapping in aquatic resource inventories)

The imagery is typically interpreted, and landform or habitat attributes are recorded in a database and/or on a map. High resolution, planimetric mapping (e.g., 1:5,000 scale mapping) is usually not possible because imagery is oblique and uncontrolled. For example, in the coastal mapping surveys, the species of bull kelp (Nerocystis) can be clearly identified on the imagery but estimates in transferring the information about a kelp bed to a planimetric map base are inherently uncertain (i.e., _10m accuracy for any single boundary of the bed).

Most recent surveys provide a burned-in GPS location of the video image, assuring that the location information cannot be separated from the imagery; in some aquatic surveys there is limited data entry provided during the survey. Older surveys, including most of the coastal surveys, provide detailed flightline maps, with fixes every 30 seconds. Although this type of flightline map provides as accurate a position of the imagery as GPS, it is not physically attached to the image and subsequent copying often results in separation of the map and tape.

Planimetric Mapping AVI Surveys

Planimetric AVI mapping involves the use of vertical AVI; that is, where the camera is pointing vertically downward such that a "horizontal" image is collected. Such an image is directly analogous to a vertical photo. The scale of the image is related to the altitude of the aircraft and focal length of the camera. Resolution of features will vary, depending on the image scale (e.g., 1:5,000 or 1:20,000), the stability of the aircraft and the image resolution (in turn a function of the camera, recorder and display system).

Only two programs in BC have conducted planimetric mapping from the AVI (see Appendix A, Projects 4 and 8). The Terra/LIDAR system (see Appendix A, Summary No. 8) uses a vertically mounted camera in conjunction with a laser profiler and an inertial navigation system. The inertial navigation data allows the precise attitude of the aircraft to be known such that imagery can be corrected to _1-2m accuracy's and landscape features can be precisely mapped. The mapping requires considerable post-survey processing and is expensive; however the mapping information is very precise and is used for input to hydrographic charts.

Vertical images can be used for mapping in an uncorrected mode (no corrections for attitude fluctuation of the aircraft) and vegetation polygon mapping accuracy's of <10% are estimated (D. Campbell, 1995, pers. comm.; see additional Range and Bearing summaries in Appendix B).

Controlled and uncontrolled planimetric mapping surveys have only been conducted at two locations within BC. Although surveyor-quality maps can be produced, the processing required is expensive and for some applications, mapping from conventional, existing vertical air photos would provide a less expensive alternative (see Ham 1995).


2.3 Costs of Survey Techniques

[Back to TOC] [Previous] [Next]

Costs of survey techniques can only be estimated, as there are a wide range of survey objectives and variables. For example, some of the survey techniques require a specialist on-board during the overflight to provide a synchronous narration to complement the imagery. A range of survey costs are provided for a variety of survey objectives in Table 2. The costs assume private-sector charge-out rates for personal. No post-survey processing (e.g., interpretative products) is assumed such that products are (a) a videotape with inflight commentary and (b) a flightline map and/or file.

The use of a helicopter creates the major difference in survey costs, although the quality of the imagery is often more consistent. In the case of coastal surveys, for example, fixed-wing aircraft are generally too fast and tend to "straighten out" the coast, leaving long shots into embayments unsuitable for biophysical mapping.

Unit costs are likely to be reduced during longer surveys as there would be less mob/demob effort per kilometre of survey. It is expected that the relative differential between survey costs would remain approximately the same for longer surveys.

Note: Cost based on 200km survey, 1 hour for non-survey flight assumed; 1 day tech time for mob/demob; products are imagery + flightline map or file


[Back to TOC] [Previous] [Next]

 

HTML Created: April 98

Copyright © 1998 Province of British Columbia

Published by the Resources Inventory Committee