How many photos are needed for drone surveying? 4 ways to decide

When considering drone surveying, it’s common to be asked, “How many photos are needed?” For the client this is important to estimate approximate cost and work time, and for the operator it’s indispensable for estimating flight plans, battery plans, and analysis load. However, the number of photos for drone surveying is not determined mechanically by area alone. It results from multiple overlapping conditions such as required accuracy, desired deliverables, site shape, elevation differences, presence of obstacles, flight altitude, and overlap settings.

For example, even for the same 1 hectare site, the number of photos needed differs greatly between a flat vacant lot where you want an orthophoto and a construction site with many slopes and structures where you want a three-dimensional understanding. Conversely, if you decide the number of photos first, you may find it insufficient and need to re-shoot, or you may take unnecessarily many photos and make processing heavy. The important point is to treat the number of photos as an outcome and to build the plan from the conditions you should determine first.

This article organizes and explains how the number of photos for drone surveying is decided from four practical perspectives. To be useful both for clients preparing request conditions and for operators planning flights, we explain why each condition affects the number of photos, what kinds of sites tend to increase or decrease the count, and how to judge concretely.

Table of contents

• Start from understanding that the number of photos for drone surveying is not fixed

• Method 1: Work backwards from deliverables and required accuracy

• Method 2: Judge increases/decreases by area and site shape

• Method 3: Flight altitude and overlap settings greatly change required photo count

• Method 4: Plan considering site conditions and re-shoot risk

• What clients should confirm when considering photo count

• What operators should check on site when considering photo count

• Common misconceptions and how to revise them

• So how should the number of photos actually be decided?

• Summary

Start from understanding that the number of photos for drone surveying is not fixed

First, keep in mind that there is no universal answer like “for this area you always need X photos.” Of course, there are approximate guidelines. If you photograph flat terrain at a fixed flight altitude with common overlap settings, you can roughly calculate the number of photos needed based on area. But in practice, cases that can be decided by area alone are actually the minority.

That’s because photos for drone surveying are not simply used to record the site; multiple photos are overlaid to reconstruct positional relationships and produce deliverables such as orthophotos, point clouds, and 3D models. Therefore, not only how each photo captures the scene but also how adjacent photos overlap, the amount of feature information in the captured objects, shadowing, slope orientation, and occlusion by trees or structures all affect whether the analysis succeeds.

A common misunderstanding is “fewer photos are more efficient and better.” Indeed, taking unnecessarily many photos increases flight and processing time. But if you fall below the necessary number, photos that should connect may not match well, leading to missing or distorted point clouds and warped orthophotos. If you have to re-shoot the site as a result, the benefit of having initially cut several dozen photos is quickly lost.

Another misunderstanding is “high-performance aircraft or cameras can make do with fewer photos.” High camera performance helps, but it cannot compensate for lack of overlap or occlusion by itself. Especially on sites with slopes, reclaimed ground, retaining walls, bridge abutments, or mixed fill and cut areas, the key is not how much information appears in a single image but whether it’s captured at appropriate angles and with sufficient overlap.

Therefore, when deciding the number of photos, first clarify “what and to what extent you want to deliver,” then combine that with terrain and flight conditions. Below we explain that decision process in four perspectives.

Method 1: Work backwards from deliverables and required accuracy

The first perspective for deciding photo count is what final deliverables you require. This is the most important point—if this is vague, the photo count will be vague too. That’s because the required photo density and overlap differ depending on whether the main purpose is an orthophoto or whether you aim for 3D shape understanding or as-built verification.

For example, if you only want an orthophoto for current-condition confirmation, a relatively simple nadir-only flight plan may suffice. On the other hand, if you plan to use the data for volume calculations, carefully track slope geometry, or verify shapes around structures in 3D, nadir images alone are often insufficient. To improve 3D reproducibility, the same location needs to appear in multiple images with sufficient overlap, which tends to increase the number of photos.

Required accuracy also has a large impact. For internal overview or before-and-after comparisons, some coarseness may be acceptable. But if you intend to use the data for quantity calculations or as-built management, requirements for positional accuracy and shape reproducibility become strict. Such projects require careful consideration not just of coverage but also of ground pixel size (GSD), overlap, and how ground control is established. As a result, the number of photos for the same area tends to increase.

Clients should be careful not to confuse “wanting photos” with “wanting survey deliverables.” The number of aerial photos to take and the plan needed to produce analyzable survey outputs are different matters. For instance, a few dozen attractive photos may be enough for a site report, but survey analysis may require many continuous captures. If the client’s specification is vague, discrepancies like “the expected accuracy was not achieved” or “the required cross-section cannot be produced” are likely to arise.

From the operator’s viewpoint, it’s important to list the requirements early for each deliverable type. Whether the focus is orthophoto, point cloud generation, or emphasis on structure sides and slope toes and crests will determine whether nadir-only shots are sufficient or whether oblique shots and supplemental flights are needed. Insufficient clarification can result in completed flights that still fail to reproduce necessary areas.

On site, it’s tempting to think “let’s just shoot at high density to be safe,” but this is half true and half risky. Excessive shooting increases processing time and data size, which matters if you’re processing multiple sites concurrently or have tight deadlines. The important point is to secure sufficient density for the purpose while avoiding unnecessary overshooting. For that reason, base the number of photos on the deliverables and required accuracy, not on area alone.

Method 2: Judge increases/decreases by area and site shape

The next perspective is the size of the target area and the site’s shape. Many people first focus on area, but in practice not only “how many hectares” but “what shape it is” strongly affects photo count.

For example, for the same area, a rectangular reclaimed lot is more efficient to cover with back-and-forth flight lines, while a narrow, elongated road, waterway, or meandering slope section requires different flight line planning. Compact areas allow efficient reciprocal flights, but narrow sites tend to have more losses at the ends; making turnarounds and providing margin widths can increase the number of photos compared to the area. In other words, simple per-area photo counts can fail to capture reality.

Sites with elevation differences are a typical example where photo count tends to increase. On flat ground you can fly at a consistent altitude and images look fairly uniform. But on sites with slopes, steps, and mixed fills and cuts, the distance to the ground changes with location even at the same flight altitude, causing variance in ground pixel size and how things appear. On steep slopes, nadir photos alone may not capture slope detail well, and even with sufficient overlap, shape reconstruction can be weak. In such cases supplemental flights or revising shooting conditions are necessary, increasing photo count.

Sites with many trees, temporary structures, heavy machinery, or material storage also need caution. For drone surveying it’s important to stably capture the ground surface; when occlusions are many, you need to view the same position from multiple directions. On construction sites, the locations of materials and machines may change daily, making areas harder to see than assumed during flight planning. Such sites are easily underestimated if you only consider area and require margin in the photo plan.

Clients often overlook that the boundary of the target area can be ambiguous in reality. Even if the construction area is clear on drawings, on site you may want to include adjacent slopes, drainage channels, or connecting roads. Edge areas tend to have unstable quality in processing, so it’s common practice to shoot beyond the necessary range rather than right up to the limit. That margin increases the number of photos. If the client does not precisely share the required extent, you can end up with “the site was captured but the necessary edge areas are missing.”

Operators should identify areas where flight inefficiency will occur by looking at site shape, not just area. Consider whether many turnbacks are needed, whether you must avoid no-fly objects, or whether multiple takeoff/landing points are required—these affect not only photo count but overall operations. Even if the flight planning software looks neat, wind direction and safety management on site may prevent you from executing the plan as assumed, so walking the site when possible is very valuable.

In short, area is an entry point for estimating photo count but not the decisive factor. Only by considering size together with shape, relief, occlusions, and edge handling do you approach a realistic number of photos.

Method 3: Flight altitude and overlap settings greatly change required photo count

The third perspective is flight altitude and overlap. These directly affect photo count and are the parts the operator can most concretely adjust.

Lower flight altitude reduces the area covered per image, but it captures ground detail more finely and makes it easier to capture small features. Conversely, flying higher covers a wider area per image and reduces the number of photos, but ground resolution becomes coarser and reproducing fine details is harder. Thus, raising altitude simply to reduce photo count can prevent meeting required accuracy or expressiveness.

Importantly, flight altitude cannot be set in isolation. It must be considered together with required deliverable accuracy, site elevation variation, safety distances to nearby obstacles, regulations, and site rules. For example, sites that require detailed checks of construction status or capturing small structures benefit from somewhat lower altitudes, but this increases the number of photos, flight runs, and battery swaps. For broad rough reclaimed sites where overall understanding is the main goal, prioritizing efficiency with higher altitudes might be appropriate.

Overlap also greatly influences photo count. Overlap is the setting for how much adjacent photos overlap forward and sideways. A certain minimum overlap is needed to stabilize processing. If overlap is too low, common tie points between photos are insufficient and images fail to connect. In areas with few ground features, repetitive patterns, or strong shadows, generous overlap is effective.

However, increasing overlap naturally increases photo count. On some sites you may want higher forward/side overlap to ensure processing stability, but this can balloon data volume. The important point is not to use “standard settings” without thought but to adjust meaningfully for site characteristics. Flat open terrain may be okay with moderate overlap, while slopes and areas near structures often need more than standard settings.

A common misconception is “higher overlap is always better.” While safer than insufficient overlap, excessive overlap can create many similar images and worsen processing efficiency. Also, increasing overlap doesn’t help if critical areas are obscured by shadows or occlusions. You can end up with more photos but still lack information for the essential parts.

From the client’s perspective, when estimating, it’s important not to think only “X hectares means Y photos” but to confirm what level of accuracy and reproducibility is expected. From the operator’s perspective, adjust flight altitude and overlap for the site and be able to explain why you chose those settings so you can align expectations with the client. Understand that photo count is the result of those settings, not an independent control metric.

Method 4: Plan considering site conditions and re-shoot risk

The fourth perspective is site conditions and re-shoot risk. Although you can theoretically calculate the minimum number of photos, unexpected things happen on site. Therefore, in practice you should plan not for the “theoretical minimum” but for a photo count that ensures quality under actual site conditions.

Wind has a large effect. Strong wind destabilizes the aircraft attitude and affects photo stability and flight line accuracy. Especially at edges or points with altitude changes, deviations can prevent the planned overlap from being achieved. Even if the plan meets the count, actual flight quality may drop, so planning with margin is safer.

Sunlight conditions cannot be ignored. Shooting during times of strong backlight or long shadows makes the same surface appear very different across photos. Depending on soil condition and slope orientation, one side can be very dark and feature points hard to detect. On such sites, changing the shooting time, planning supplemental captures, or anticipating re-flights on other days can be more efficient than trying to shoot the minimum number at a bad time.

On construction sites, movements of people, vehicles, and heavy machinery are also important. Moving objects can change what appears in each photo and disturb processing. Even the addition of material storage or temporary structures changes shooting conditions. For these sites, aligning construction timing and shooting timing can be more important than strictly adhering to theoretical photo counts. Bad shooting conditions negate the effect of photo count on quality.

How to think about re-shoot risk is important. Clients often want to minimize cost and opt for as simple a plan as possible. However, if you tighten photo counts to try to complete in one pass, small condition changes can force re-shoots. Re-shoots incur costs beyond photo count: equipment prep, travel, supervision, safety management, and rescheduling processing. Therefore, a plan with appropriate initial margin tends to stabilize overall cost.

Operators must judge how much margin to include per project. On flat vacant lots with few obstacles and stable conditions, relatively tight plans can work. But sites with elevation differences, many slopes, moving machinery, nearby power lines or trees, or limited access routes are safer if planned with supplemental flights or backup captures. This “judgment on margin” is where practical experience often shows differences.

What clients should confirm when considering photo count

The number of photos for drone surveying is not only an operator’s issue. If the client fails to organize information, photo count estimates can vary widely. Especially before ordering, what you confirm can greatly change later rework.

First clarify what deliverables you want to receive. The shooting policy changes depending on whether you want an orthophoto, a point cloud, or a simple 3D model. Also share which areas are important. Whether you want an overall view of the reclaimed land, a focus on slopes, or detailed capture around structures affects how photos are allocated.

Next, organize the required extent not only on drawings but also from a practical standpoint. Edge margins are needed on site and you may want to see interactions with surrounding areas. If you communicate too small a required extent at the ordering stage, the processing results may miss the parts you want. Conversely, distinguishing between must-have and reference areas helps improve shooting efficiency.

Also share site conditions. Elevation differences, trees, heavy machinery, temporary structures, working hours, access restrictions, and safety rules all directly affect the photo count discussion. Asking “how many photos can you do?” without these details forces answers with wide ranges. If you want a more realistic estimate, providing more site condition information is the quick route.

Clients centering their contract on photo count alone is not advisable. What matters is whether you can obtain deliverables at the required quality. Photo count is a means to that end and should be considered together with deliverable quality.

What operators should check on site when considering photo count

For operators, photo count is an indicator for both flight planning and processing planning. Therefore, don’t decide solely by desk calculations—be aware of what you should check on site.

First, look for areas that may be hard to see even if they look fine from straight above. The backs of slopes, adjacent to structures, around tree lines, and around retaining walls often lack information in nadir photos. If you identify these areas in advance, you can plan supplemental shots.

Next, check the safety of takeoff/landing points and flight routes. If safe routes are limited, ideal grid flights are hard to execute. Power lines, trees, work vehicles, and pedestrian flow can reduce planning freedom and lower flight efficiency, increasing photo count. Prioritize safety over compact efficiency.

Also consider post-processing workload. Even if you meet the required number of photos, over-shooting increases data organization and processing time. Consider delivery deadlines, workstation performance, and processing environment when targeting an appropriate photo count. On site, keeping in mind “insufficient is bad but excessive is also costly” helps balance decisions.

Common misconceptions and how to revise them

There are several common misconceptions about photo count in drone surveying. Clarifying them reduces mismatched judgments.

First: “Wider sites always require more photos.” This is a tendency, but small sites with elevation change and many occlusions can require more photos per unit area than flat open sites. Area is important, but not sufficient by itself.

Second: “High-resolution cameras allow reducing photo count.” While high resolution helps, it cannot fully compensate for lack of overlap or viewpoints. Survey analysis requires not only detail in single images but consistency across multiple images.

Third: “Taking many photos always improves quality.” Insufficient photos are a problem, but more photos alone are not the solution. A large number of similar images only slows processing; if necessary areas aren’t captured from the right angles, quality won’t improve. Capture necessary areas under the necessary conditions—this is key.

Fourth: “The number calculated by flight-planning software is exactly what’s needed.” Planning values are a good starting point, but wind, obstacles, access restrictions, sunlight, and machinery layout change on site. Don’t accept plan values uncritically; adjust based on site conditions.

So how should the number of photos actually be decided?

As shown so far, the number of photos for drone surveying is not determined by area alone. In practice, first organize the deliverables and required accuracy, then confirm the target area and site shape, adjust flight altitude and overlap, and finally plan considering site conditions and re-shoot risk. Thinking in this order naturally narrows down the number of photos.

Conversely, deciding “how many photos” first makes it easy to omit necessary checks. Clients should evaluate plans not by photo count alone but by whether the plan is appropriate to obtain the required deliverables. Operators should consider not only desk efficiency but whether the plan will work on site and whether it’s realistic up to processing.

Photo count is neither a sign of excellence when small nor reassurance when large. Finding an appropriate value according to purpose, terrain, settings, and site conditions leads to both quality and efficiency. Especially for surveying purposes, the important thing is not the number of photos taken but whether the shooting method meets the required accuracy and reproducibility.

Summary

The number of photos for drone surveying is not fixed but depends on conditions. The main decision axes are deliverables and required accuracy, area and site shape, flight altitude and overlap, and site conditions and re-shoot risk. For the same area, the number of photos varies between a flat vacant lot and a construction site with elevation differences, and whether the primary goal is orthophoto or 3D understanding also changes the plan.

Clients should clarify “what and how much you want to capture,” not just “how many photos.” Operators should not rely solely on the flight-plan photo count but should build realistic plans that account for wind, obstacles, sunlight, and work conditions to reduce re-shoots. Correctly judging the required number of photos is important not only for cost control but also for quality assurance and avoiding rework.

Also, while drone surveying excels at wide-area aerial capture, combining ground-based reference checks and supplemental ground measurements can improve practical usability. For example, if you want high-precision positional confirmation at key points or supplementary ground positioning, using an iPhone-mounted GNSS high-precision positioning device such as LRTK and combining drone-acquired overall data with ground positioning information can be effective. Thinking in terms of role sharing between air and ground helps organize the approach to photo counts and makes it easier to create drone surveying plans suited to practical use.