Tea plant-gap inspection using drone orthomosaics

Start with the operational question

Define a gap according to the management question and inspect its relation to the tea rows, paths, drains, and pruning areas. Use imagery with sufficient detail for that definition.

Separate observation from explanation

A dark opening can be a shadow or recently pruned patch, and a green patch may contain weeds rather than tea. Visibility and class uncertainty should remain attached to every candidate.

Define the object before counting it

Describe exactly what qualifies as one object. A visible crown, a live plant, a productive plant, a planted position, and a gap are different definitions. Set rules for overlapping objects, partial objects at the boundary, ambiguous shadows, and objects below the useful image resolution. Record these rules with the dataset so that another reviewer can reproduce the interpretation.

Evaluate both false detections and missed objects. Inspect a reference sample that includes locations where the method proposed nothing, since reviewing only the proposed objects cannot reveal omissions. Cover contrasting ages, canopy densities, illumination, image quality, and boundaries. Keep the reference interpretation independent from the automated output where possible.

Attach an image date, method version, review status, and field status to each candidate. Retain uncertain cases rather than forcing them into confirmed categories. When reporting totals, explain the counted population and the areas excluded from reliable interpretation. A corrected inventory should preserve the earlier result and the reason for each accepted, rejected, merged, or moved object.

Treat pruning history as essential map context

Tea monitoring should begin with a block register that includes cultivar or planting material where known, planting age, pruning history and the current plucking cycle. Tea is managed for repeated vegetative harvest, so canopy changes can reflect planned operations as well as unwanted stress. A recently pruned block should not be ranked against a block with a developed plucking surface without explaining the difference. Retain those management classes in tables, map legends and field forms.

TNAU's tea guidance discusses well-drained soils, organic matter, pruning and plucking as parts of the production system. These topics identify useful observations; they do not establish universal treatment rates for every tea-growing region. Record slope position, drainage condition and visible soil exposure alongside crop observations. On hillsides, shadow and illumination can change the appearance of adjoining rows, so compare image patterns with terrain and the direction of sunlight before assigning a biological interpretation.

Use field observations that explain the plucking surface

A field visit can document shoot condition, canopy continuity, gaps, weeds and the recent management operation at each sampled location. Define the observation unit and use it consistently: a photograph of one bush and a judgement about an entire block are different scales of evidence. Select both apparently normal and affected places within comparable pruning and plucking situations. Record the date and time because harvesting activity may change the surface between an image acquisition and the visit.

Where the concern is moisture or drainage, inspect the ground and access routes as well as the leaves. A canopy signal alone cannot distinguish inadequate water supply from poor root conditions or another constraint. Where the concern is nutrition or disease, use an agreed sampling or diagnostic procedure and retain the resulting laboratory or specialist record. Avoid converting a colour class into a fertiliser amount or plant-protection recommendation without the necessary agronomic evidence.

Connect the field map to operations and quality records

Repeated harvesting makes the timing of a comparison particularly important. Keep plucking dates, relevant labour or machine operations and collected leaf quantities associated with the correct block and reporting period. If a factory intake combines leaf from several blocks, preserve the aggregation rule instead of presenting the intake as a measurement of one small mapped patch. Tea quality also involves harvesting and processing decisions that are not visible in a satellite image.

For the next monitoring cycle, revisit the same observation locations and note changes in pruning status, canopy development and drainage. Evaluate whether an alert correctly identified a place needing inspection and whether the follow-up produced useful evidence. A stable colour pattern can be operationally important even when its cause remains uncertain; an abrupt change can be routine if it coincides with a documented intervention. Publish the event history with the map so managers can distinguish those situations and plan the next inspection deliberately.

Match image detail, timing, and processing to the task

Describe the smallest feature or change that the image must resolve. Field-level vegetation monitoring and individual-plant inspection require different spatial detail, and neither requirement is established by the displayed zoom level. Check native band resolution, actual acquisition date, and the amount of usable coverage. A resampled image can have more pixels without containing more independent detail.

Review processing choices before comparing dates or sources. Surface reflectance, raw radiance, visual photographs, and styled map tiles are different products. Note atmospheric treatment, radiometric calibration where relevant, image alignment, and quality masks. Where sources differ, compare at a scale that both can support and explain whether the comparison concerns spatial pattern or numerical agreement.

Keep a field reference appropriate to the question. A photograph may confirm that a stand gap exists, while a calibrated instrument or laboratory method is needed for a quantitative property. Schedule field observations close enough to the image to represent the same condition, and document the delay. If the crop changed between acquisitions, retain that event as part of the interpretation instead of treating the sources as simultaneous.

Design reference observations that represent the field

State the population that the sample should represent and the question it must answer. Select locations across relevant crop stages, terrain, management, and image conditions. Include ordinary areas and difficult cases, not only conspicuous anomalies or convenient roadside stops. Record selection rules so that later users understand where the reference is representative.

Define the observation unit, area or depth, timing, instrument or laboratory method, and required replicates using the appropriate discipline's protocol. A point observation and a satellite pixel have different spatial support. Record enough context to evaluate that mismatch. Keep inaccessible sites in the sampling log and explain any replacement rather than silently dropping them.

Separate reference collection from the result being evaluated when feasible. Review observer consistency and missing data before calculating agreement or error. Retain raw observations, corrections, and exclusions with their reasons. More samples can improve precision, but a large biased sample still fails to represent excluded conditions. Use the results within the population and period that the sampling design supports.

Define the deliverable and approval

Prepare a candidate gap map and a field confirmation form that record actual crop absence, management constraints, and any approved corrective action.

A practical review before sharing the result

Check the observation against its source and make sure the field identifier is correct. Review dates, units, missing values, quality exclusions, and geometry versions. Ask a colleague to reproduce one representative result from the recorded inputs. This small independent check often exposes a misunderstood field code or a processing assumption before it affects an entire report.

Write down the alternative explanations for the mapped pattern and identify what evidence would distinguish them. Keep a measured observation separate from the interpretation and from any treatment or management instruction. A decision should name its responsible reviewer and account for local operating procedures. If the evidence is insufficient, the appropriate output is a targeted inspection rather than a confident claim.

Keep the original result, any reviewed version, and a short change log. State which locations and periods the output can reasonably describe. When sharing a map, provide a clear legend, source dates, and an accessible explanation of uncertainty. The recipient should know whether a feature is a confirmed observation, a candidate awaiting inspection, or a model-based estimate.

Turn the investigation into a repeatable field cycle

Prepare the visit with a map that can be used at the intended location and a concise list of questions. Confirm access, permissions, equipment, and the availability of a person familiar with the field. Choose representative stops as well as unusual areas. Record where the planned route could not be followed, because missing observations can change how the result should be interpreted.

After the visit, review the observations while the team can still clarify them. Connect photographs, measurements, and comments to the same field and visit identifiers. Assign a follow-up status and preserve unresolved cases. A correction should explain what changed and why; it should not erase the original evidence or make an earlier interpretation appear to have been certain.

At the next suitable observation date, compare conditions using the same definitions and review any management or weather events that occurred between visits. Record the action actually completed and its operational outcome. This creates an evidence trail from an initial pattern to field confirmation and later review, making the mapping process easier to improve over successive seasons.