-2

It is on the need of accuracy assessment of classification product.

Land cover/use classification is a prominent analysis in remote sensing. Now the question is if I generate a landcover from LANDSAT, do not perform ground survey for cross-checking and want to do a scientific analysis with this classification, is it scientifically invalid?

If I do this cross-checking with some very high-resolution image (5m) and Google Earth, is it robust method to fit scientific standards, if yes then how much.

Now my question is can anyone list some articles link that used their research based on un-ground-truthed image classification data or at least guide me on how to find those type of article

By ground survey, I mean field survey in body with some instruments like dgps, densitometer etc.

I asked Accuracy of the Mathew Hansen Dataset? that has some tone matches with this question but still feeling unhelped. I emailed Mr. Hansen about a month ago but no response yet.

Improve this question
6
  • 1
    Generally speaking, it is better to edit the unanswered question than to add another vague question to the pile.
    – Vince
    Commented Nov 7, 2016 at 12:22
  • OK edited the question and narrowed the asked point.
    – Learner
    Commented Nov 8, 2016 at 4:52
  • Asking for a list of articles leaves this question as too broad. Likewise requesting that someone teaches you how to put together such a list.
    – PolyGeo ♦
    Commented Nov 8, 2016 at 5:40
  • @PolyGeo could you please suggest how to solve the problem- I want to develop a scenerio by mix of hansen data with some other secondary data but this scenerio may not match the proper scientific rigor on the sole ground of no accuracy assessment done for hansen data at subnational level.
    – Learner
    Commented Nov 8, 2016 at 5:57
  • "is it scientifically invalid in academia?" sounds like a question for the Academia Stack Exchange.
    – PolyGeo ♦
    Commented Nov 8, 2016 at 6:06

1 Answer 1

Reset to default
2

It is very common for a published study to use the remotely sensed datasets as the reference data in validation instead of field data, especially when the study area is large (eg continental or global scale). There are a lot of reasons for not using the field data, including:

  1. no data available,
  2. data does not meet the validation purpose (eg poor in quality or lack of consistency etc)
  3. collecting field data is too expensive (eg for remote locations such as Siberia) or impractical (eg for large scale studies).
Improve this answer
4
  • Emm,I think I have got 3rd reason most suited here.But could you give light on using hansen data for sub national level and deriving analysis based on it for scientific knowledge.I mean analysis usin hansen with other data to dig into developimg a scenerio..will this research be discarded on the ground of not robust.
    – Learner
    Commented Nov 8, 2016 at 3:59
  • "will this research be discarded on the ground of not robust" sounds like it may be better researched at the Academia Stack Exchange.
    – PolyGeo ♦
    Commented Nov 8, 2016 at 5:41
  • @PolyGeo I am stumbling on this question for many days:( I asked this question in the academia that followed a discussion and concluded into the reason of discarding the research on the ground of not-so-far-done-accuracy and finally they asked to get this answered here.
    – Learner
    Commented Nov 8, 2016 at 5:53
  • The Q&A from Academia sounds like it should be linked to at the beginning of your question here because that triggered it.
    – PolyGeo ♦
    Commented Nov 8, 2016 at 6:09

Not the answer you're looking for? Browse other questions tagged or ask your own question.