Month: March 2022

How to Select a Mapware Pricing Plan

A guide for selecting the drone photogrammetry price plan that works best for your business

Last September, Mapware introduced a Pay-As-You-Go pricing model in which customers could pay 2¢ per megapixel (MP) to process drone imagery in our photogrammetry engine. This was a great, no-strings-attached way for individual drone pilots, infrequent photogrammetrists, and new entrants to the field to try out our software. But we’re optimistic that many of these Pay-As-You-Go customers will eventually scale their businesses up, and we wanted to give them higher-value plan options for their increased processing volume.

To that end, we’ve added three new subscription-based plans tailored for different-sized customers. Read on to figure out which one is right for you.

The plans

So, what plan is right for you? Below, we’ll describe a type of customer who benefits from each plan tier. We’ll also spell out the precise limits of each plan so you know when it’s time to upgrade.

Pay-As-You-Go

Best for one-time projects, infrequent drone mappers, and new pilots

If you are considering UAV mapping as a new income stream, our Pay-As-You-Go plan gives you a no-strings-attached way to test the waters. It’s also perfect if your business does photogrammetry infrequently or is planning a one-time project.

There are two reasons:

First, calculating your project’s processing cost is straightforward; you just multiply the number of photos by the size of your camera to get the total megapixels, and then multiply that by $0.02 to get your bill. For example, the DJI Mini 2 has a 12 MP camera. If you take 200 photos of a site, your cost will be (12 x 200 = 2,400 x $0.02) = $48.

Second, the lack of a subscription cost frees you from having to process models at a certain frequency to justify a monthly or annual bill.

Solo

Best for the serious single-pilot operator

There’s an easy way to decide between the Pay-As-You-Go plan and the next level up, Solo. The magic number is 4,950 MP per month.

We offer the Solo plan to give individual UAV mappers a discount for their increasing success, and that discount starts once you are regularly processing more than 4,950 MP per month. At 2¢ per megapixel, 4,950 MP would cost $99 under the Pay-As-You-Go plan.

The Solo plan also costs $99 per month when paid annually. But upgrading to Solo entitles you to process up to 500,000 MP per month. That should keep you pretty busy for a while.

Squad

Best for your growing flight ops department

As a solo UAV mapper, you’ll be hard-pressed to exceed 500,000 MP per month. But if you attract more business than one pilot can handle, you’ll want to hire additional staff to avoid turning down work. The Squad plan is designed for this. It lets you share your monthly processing limit within a team of up to 10 pilots.

With more pilots, you might see a significant increase in your monthly photogrammetry processing. To accommodate this, the Squad plan also doubles your monthly storage limit from 5 terabytes (TB) to 10 TB and your processing limit from 500,000 MP to 1,000,000 MP.

Fleet

Best for enterprise photogrammetry

The Fleet plan is for established photogrammetry businesses that want to tailor their plan’s capacity. Call us to upgrade from Squad to Fleet if you want to do any of the following:

  • Enable support for a sensor type other than the standard electro-optical (EO)
  • Share the plan with more than 10 pilots
  • Store more than 10 TB per month
  • Process more than 1,000,000 MP of images per month

Reading the pricing page

Next, here are a few key points about our new pricing page for anyone who finds reading these sorts of things difficult.

Plan differences

Even though it looks complicated, there are only a few key differences between each plan. If you’re choosing your first plan or considering an upgrade, focus on the following:

  • The amount of imagery you need to process per month, in megapixels (MP)
  • The amount of free cloud storage space you need per month, in terabytes (TB)
  • The number of Mapware users you need on the plan

Plan similarities

On the other hand, all Mapware users get certain things for free. No matter what plan you select, you will get an unlimited number of the following:

  • Images per model
  • Models per project
  • Ground control points (GCPs)
  • Projects
  • Uploads and downloads (ingress and egress) of all project files

A note on “images processed per month”

On our pricing page, the Compare Features table has a row listing the “images processed per month” covered under each plan. It’s important to note that this is not a hard limit. It’s merely an estimate of how many images you could process before reaching the plan’s MP processing limit if you used a 20 MP camera.

For example, under the Solo plan, we estimate that 500,000 MP lets you process about 25,000 images per month if you capture them using a 20 MP camera. But using a 12 MP camera would result in image files with smaller MP values. You could process more than 41,000 images before reaching the monthly limit of 500,000 MP.

Paying annually or monthly

Finally, you may have noticed that the Solo and Squad plans have the phrase “per month with annual plan” under their price tags. If you don’t want to agree to an annual subscription, you can elect to pay for these plans on a month-by-month basis. The cost is slightly higher, but you can end the plan before the year is through.

Mapware’s Photogrammetry Pipeline, Part 5 of 6: Fusion

In the fusion step of the photogrammetry pipeline, Mapware merges all its depth maps together into a dense 3D point cloud and orients it in real-world space.

Photogrammetry pipeline so far

In the previous pipeline step, Mapware used spatial estimates from its sparse 3D point cloud along with metadata from each raw image to generate depth maps showing the elevation of each image pixel. This vastly increased the density of Mapware’s 3D data.

The purpose of fusion

Now, in the fusion step, Mapware takes advantage of its new data to generate a far more-detailed point cloud, filling in the spaces between keypoints. Mapware also georegisters the point cloud (orienting it in real-world latitude/longitude space). The result is far more accurate than anything previously in the pipeline and can serve as the basis of the final 3D digital twin.

Mapware’s fusion process

Dense point cloud generation

First, Mapware brings together the complete list of raw images and depth maps. It also brings together data from all previous pipeline steps including keypoints, image pairings from homography, 3D point estimates from SfM, and pixel elevations from depth mapping.

Then, Mapware runs this data through its stereo fusion algorithms, which project each depth map’s pixels into a temporary 3D space to generate a new point cloud. The code then tries to establish the accuracy of each point by comparing depth maps. If each point is described in at least three depth maps, consensus is reached and the point is considered valid. Otherwise, the point is discarded.

The use of depth map pixels rather than keypoints means that the result is far better than the one generated during the SfM step. Not only are there many more points with fewer gaps between them, but those points are in far more-accurate spatial positions. This is referred to as a high-resolution dense point cloud.

A comparison between sparse and dense point clouds generated from photos of a car
The image on the left is a sparse point cloud created during SfM, and the image on the right is a cleaner and more-complete dense point cloud created during fusion.

Georegistration

Next, Mapware fits the high-resolution dense 3D point cloud into a real-world coordinates system.

To do this, Mapware reads each raw drone image’s Exchangeable Image File Format (EXIF) metadata to determine the camera’s GPS latitude, longitude, and altitude at the time of each shot. It also retrieves the relative camera positions it calculated earlier during the SfM step. Next, Mapware determines the best linear transformation between those two data sets. With this calculation, along with the relative position of the points in the new dense point cloud, Mapware can determine the position of each point within a real-world coordinates system.

Now that it’s tied to physical coordinates, the dense point cloud can serve as the basis of Mapware’s final 3D digital twin.

Next steps

At this point, Mapware has generated a viable 3D model of the environment. But it is not in a format that many photogrammetry customers want. In the final step of the photogrammetry pipeline, structured output, Mapware will generate additional output formats from its 3D model that users can choose from. We’ll discuss that final step next time.