Upkeep - Visual inspection

From Wikicap - European Commission



This visual inspection, possibly followed by a Upkeep_-_Congruency_test determines if a land change anomaly does indeed exist and represent a manifest change. It offers a quick and simple stepwise method for inspecting LPIS reference parcels for the occurrence of such manifest change.

The main outcome of a visual test is to verify if there is a manifest change. This test has three possible outcomes:

  1. The parcel has been stable and there is no need for a remedial action.
  2. The test's outcome is inconclusive and the parcel is subjected for a congruency test if a change was reported externally.
  3. The parcel has indeed changed and remedial action is needed.

Note: This visual inspection procedure can "mutatis mutandis" also be applied for application during the OTSC, measuring a agricultural parcel that wholly occupies one reference parcel. (see 2.6.1 of the art34 guidance)


Activity Diagram


Ensure all issues of this checklist are met:

  • LPIS, past and actual orthoimage are all set up to ensure good display functionality.
  • The visual test conditions are the same as for delineation:
The visual scale primarily depends on the image resolution and the MMU for the features concerned, parcel size and local ground conditions. A case by case assessment is needed, however, within the range of imagery used for LPIS upkeep:
  • that visual scale should always be larger than 1:5,000
  • the largest recommended scale is 1:1,000 (i.e. a slightly larger scale is possible if the resolution and local conditions allow)
The decision thresholds defined in this guidance are, unless specified otherwise, not dependent on this visual scale.
  • If possible, any systematic shift of image is detected and locally corrected. Use the procedure to determining distance and direction of the shift.
  • The RP land can be displayed together with parcel perimeter in LPIS (RP) on both past and present orthoimage.
  • The operator is familiar with the specific ancillary data and metadata and can access these during the inspection:
  • the life cycle of the vegetation
  • the boundary delineation precision of the LPIS
  • an indication of the scene view angle or approximate camera position.
  • an estimation of distance and area on the working scale.


BP0105001 Visual inspection.png

diagram 3ci.1: visual inspection activity diagram

During visual inspection, the following questions should be answered in sequence. The test procedure is completed as soon as a positive reply (“yes”) occurs. The inspection and its steps lead to qualitative answers. They do not require digitization, only an assessment by a “trained eye”.

  1. Do I experience problems in displaying the RP perimeter on the locally geocorrected current image? (e.g. does the joint display make obvious sense?)
  2. Do I experience problems in viewing the whole perimeter visible on the orthoimage (e.g. nothing is blocked by clouds)?
  3. Has any of the perimeter/border features been removed, changed or displaced (> 5m) on the new image?
  4. When I toggle the old and the new image, do I observe the appearance of a new building, road or artificial object, etc. on the parcel?
  5. When I toggle the old and the new image, do I observe changes from agricultural to non-agricultural features that are located completely inside or along the perimeter of the parcel that represent an estimated ground distance of 5 m or more?
  6. When I toggle the old and the new image, do I observe land conversion changes that have occurred on the perimeter, even when appropriately taking into account:
  • any displacement due to image processing (max 1-2 pixels)
  • any displacement due to the life cycle of vegetation (caused by temporary growth processes)
  • any relief displacement of above ground and below grand features (estimated for the image viewing angle and height/depth of the feature)?
7. Identify four, well identifiable ground level perimeter points in each of the four cardinal directions (N-E-S-W). Does any of these show a displacement (RP vs image) above 1.25 m?
8. Can I confirm my replies to the above questions (with ancillary data if needed)?


The outcome depends on the first occurence of a positive reply:

  • If all questions but question 8 are answered with “no”, the parcel has not been subject to change. No remedial action is needed.
  • If question 1 or 2 is answered with “yes”, a full visual inspection is not feasible with the given image. Make the necessary provisions to inspect the parcel appropriately and in due time.
In cases where an anomaly has been reported (and a visual inspection is not feasible in the foreseeable future), a dedicated field visit should be considered to confirm the anomaly and collect appropriate data for updating the reference parcel.
  • If question 3, 4, 5, or 6 is answered with “yes”, a manifest change has been detected and confirmed. You should register its type and proceed with the upkeep procedure.
  • If question 7 is answered with "yes"
  1. the visual inspection is inconclusive in case an anomaly had already been reported by other means than this visual inspection. A congruency test should follow.
  2. Without such anomaly from external sources, the parcel can be considered unchanged.

Record the outcome of the test and the date when the visual test was performed on this anomaly.

Go back to the main upkeep article