A convincing image can still contain measurements that are unsuitable for a particular task. Quality control asks whether the inputs, transformations, and outputs meet the requirements of the intended application. This perspective follows evidence through the workflow, from acquisition metadata and reference observations to the final interpretation. It encourages teams to define acceptance conditions before seeing the result, rather than adjusting them to favour a preferred conclusion. The purpose is a repeatable way to distinguish usable observations, observations requiring qualification, and observations that should be excluded from a decision.
This working guide considers testing whether a remote sensing model transfers across mine restoration areas. The immediate decision is to target restoration follow-up visits. Useful reference evidence includes treatment records, surface condition and representative vegetation checks. The case is an illustrative planning context, with assumptions to check against the actual sensor, landscape and project requirements.
Identify what changes between domains
A model trained on one region or season learns relationships present in its development data. Applying it elsewhere assumes those relationships remain useful. New crop varieties, forest types, soil colours, management practices, illumination, or sensor characteristics can violate that assumption. Define the intended transfer explicitly: a new district, another year, another sensor, or a different class distribution. Each change creates a different test. A model that transfers across nearby fields has not thereby demonstrated reliable performance across climates or observation systems.
Keep the measurement pipeline consistent before attributing differences to the model. Reflectance scaling, band order, cloud masking, spatial resolution, composite windows, and missing value handling can all change predictions. Sensors with similarly named bands may have different spectral responses. Harmonization can reduce some differences but does not establish equivalence for every target. Examine predictor distributions, missing data rates, and examples from the new setting. Flag inputs outside the conditions represented during training so that extrapolation remains visible.
Test the destination, then adapt carefully
Design evaluation splits around the intended deployment. Hold out geographic regions to test spatial transfer and later years to test temporal transfer, keeping neighboring samples or repeated observations of the same site together when needed. Report performance for meaningful destination groups rather than only a pooled score. Class prevalence affects some metrics, so interpret changes alongside the underlying confusion matrix or residual distributions. A random pixel split can look excellent while providing little evidence about performance in an unseen region.
If local labels are available, reserve a destination assessment set before fine tuning or recalibration. Use the remaining observations to compare straightforward baselines with more complex adaptation methods. Check whether improvement comes from genuine transfer or simply from adding representative local data. Retain an option to withhold predictions when input quality or novelty makes them unreliable. Document the training domain, destination evidence, preprocessing versions, and conditions where use is unsupported. Continue monitoring once new observations arrive, since seasonal and environmental shifts can emerge after an initial evaluation. Transferability is a measured property of a particular model, workflow, target, and destination; it should not be inferred from architecture name or strong performance on the original test set.
The setting: Mine restoration areas
A mine restoration area can contain reshaped ground, drainage features, exposed material, planted vegetation and naturally developing cover. A remote-sensing product should distinguish these components and identify the outcome it is intended to assess. Mapping the extent of visible vegetation is different from evaluating restoration success, soil condition or stability. An increase in greenness may be useful evidence, but it cannot establish all of those outcomes on its own.
Treatment boundaries and dated operation records help interpret the image sequence. Compare suitable seasonal periods and account for the influence of soil background, terrain illumination and remaining industrial surfaces. Some planting patches may be too small to separate reliably in the chosen imagery. Retain observation dates, quality masks and the area contributing to each summary so that the result does not imply finer detail or more continuous monitoring than the data provide.
Field validation should record vegetation establishment, composition and the surface attributes required by the restoration objective. Include different treatment areas and apparently successful locations as well as obvious gaps. A model for a particular outcome needs reference measurements and evaluation matched to that claim. A practical first service can provide an auditable record of visible cover development and inspection priorities, while more comprehensive conclusions combine that record with the field evidence needed to assess the intended restoration performance.
Validate land-cover transitions
A mapped change should describe a transition between meaningful surface states, not merely a difference between two classifications. Begin by defining the land-cover categories, the geographic unit, and the kinds of transition relevant to the application. Categories should be interpretable in the observations available at each date. A distinction that cannot be recognised consistently can create apparent change even when the underlying surface remains similar.
Evaluate the observations supporting both sides of the transition. Differences in acquisition conditions, geographic alignment, seasonal vegetation, shadows, or class definitions can influence the result. Review stable locations alongside apparent changes, and retain the source imagery or other evidence needed to investigate ambiguous cases. Where multiple changes occur within a geographic unit, describe how the workflow represents mixtures rather than forcing a single label that conceals important variation.
Use reference evidence that reflects the timing and scale of the claimed transition. A recent inspection may confirm the current surface without establishing when a previous condition changed. Historical imagery, dated records, and complementary observations can strengthen that sequence where available. Report uncertainty in the transition itself, including ambiguous starting or ending states. This produces a change product that users can examine as a documented account of land-cover evolution rather than an unexplained set of coloured differences.
Design an evaluation that reflects deployment
A model evaluation should resemble the conditions in which the model will be used. Start by identifying which relationships the development data share: nearby locations, repeated observations of the same unit, common collection teams, or measurements from a single period. These relationships can make records less independent than they appear. An evaluation that ignores them may test the ability to recognise familiar conditions rather than the ability to operate in a new setting.
Choose a split strategy that matches the intended transfer. Holding out locations, seasons, operational units, or acquisition conditions can answer different questions. Document which question each split is designed to address, and keep the held-out evidence separate while selecting features and tuning the method. Repeatedly inspecting the same evaluation set to improve a model gradually makes that set part of development, even if it retains its original label.
Report results at a scale that is meaningful to the user. A strong average can hide poor performance for important subgroups or unusual conditions. Examine error patterns and retain examples of both successful and unsuccessful predictions. Where the evidence covers only a limited range of conditions, describe that range explicitly. The evaluation then becomes a practical guide to suitable use, further testing, and the collection of additional reference observations.
Bring the evidence back to the decision
Quality control is most useful when its results travel with the information being delivered. Preserve the reasons for accepting or rejecting an observation, and expose limitations that could change the user's interpretation. Review recurring failures as clues about the design of the workflow. A clear quality record supports consistent comparisons between operators and processing versions, while helping the team decide whether a problem originates in acquisition, preparation, interpretation, or the underlying observation requirements.
Sources & further reading
Explore the bhumisat spacecraft concept and our vision towards 2030, or return to the journal.



