Disagreements when applying Google Street View to decode aerial data

Just out of curiosity, why does it matter if a roof is decoded using the gutter or ignoring the gutter? I don't know anything about decoding ortho imagery but I'm having a hard time imagining a scenario where it would be important to know if the line work was drawn using the gutter or just the roof. Who would it cause a problem for if they made the incorrect assumption about the gutter being included?

The simple answer is that the USA is not Israel or Europe and what you've been accustomed to is not relevant here. I'm not aware of any specific statutes, administrative rules, and/or ordinances that regulate the scenario that you describe. You're not alone in being frustrated that so few things are standardized in the USA, but the sooner you accept it and stop trying to fit a square peg into a round hole, the sooner you can begin learning a new method of data acquisition and QC and increasing your value as an employee.

As a manager, my tolerance for employees that ignore my instructions and insist on doing things the way they're used to doing them is reaching an all time low. Once an employee has worked with me awhile and shown me that they can follow simple instructions, I become much more open to their suggestions regarding how to improve our work product. Don't put the cart before the horse, listen to your boss for now and know that in the future you'll likely be able to implement portions of your hard earned knowledge.

For a field surveyor, it is much easier to measure the gutter and offset its width.

Then offset the eave and you get the walls.

I would not trust StreetView as an absolute truth - around here the imagery is five to six years old

Who would it cause a problem for if they made the incorrect assumption about the gutter being included?

Oh you sweet innocent. I wish I was too.

Four cities in a county:

1. Setbacks include all overhangs, including eaves (not including gutters).
2. Setbacks are to foundations, as long as eaves are not more than 2 feet. (Do not need to show eaves.)
3. All portions of the building, including gutters must not encroach on the setbacks.
4. Setbacks are to foundations, but you must measure and verify the eaves and show them to be 2 feet or less of encroachment.

Now ask me about impervious surfaces, what to locate, whether or not a full asbuilt is needed after improvements, etc. LOL

They recently "updated" building codes, now the edge of a gutter or eve can't hang over any sideline easement. Contractors are making narrow eves and gutters to conform, guess how that will turn out.

Are you guys really using orthophoto data to determine whether or not a setback is violated? There is zero chance I would rely on orthophoto data for something that is approximately .5' such as the gutter on a roof. I'm certainly not innocent but maybe too cautious if others are making it a practice to use orthophotos to determine setback violations.

Good afternoon, thank you for your participation in the discussion. I am a bit confused that you only paid attention to “gutters”, as cross-comparison with other mapping services for 2D and 3D planimetric mapping is 2nd most important after engineer's meticulousness in my opinion.

I will explain: for the last 3 years, I have done 500-550 projects in the USA on 2D, 3D decoding of aerial data. Usually, the dataset consists of (DEM, DTM, DSM or LiDAR, orthoimage), and I very often had to decode orthoimage, DEM, DTM with 10%, 20%, 25% of the total project area with orthorectification artifacts (stitching artifacts, geometric distortions, shadow or gap areas). There were cases when we had to decode orthoimage with red tint (this is when a UAV pilot does aerial survey at sunset), or for example, we often have to deal with large orthoimage projects with very, very bad detail.

Usually, the orthoimage detail is affected by the UAV flight altitude, but unfortunately, many pilots often abuse this parameter in order to capture more area for aerial survey. When orthoimage has poor detail or has a large number of orthorectification artifacts, it is difficult and sometimes impossible to accurately identify and classify objects according to ALTA/NSPS standards. I did my own research, and if we refuse cross-comparison with other mapping services, we get from 5 to 20% of objects that are classified incorrectly.

I will give a couple of examples (actually there are many more). Example 1: there are cases when due to poor detail, orthoimage cannot distinguish the type of fence (metal, wood, concrete, wood on a concrete base), but the outline of the fence is visible, and it can be decoded by checking on Google Street View, Bing Streetside, Bing Birdseye; the type of fence is clearly recognized. I agree with “jimcox” that Google Street View, Bing Streetside do not always contain up-to-date data, but it is an indispensable source of auxiliary information, and for 3 years of constant work on such projects, you develop a visual memory. When you analyze shadows, the height of the object, and its location, you can, with a high probability, correctly classify it.

Example 2: when performing cross-comparison on Google Street View, manholes of public utilities were identified, but because of poor detailing on orthoimage, the human eye almost does not distinguish them. That is, on orthoimage, they are there, but even an experienced person is likely to miss them.

My position is that an engineer should get the most out of aerial data, but I think it's inappropriate not to use mapping services because, on large projects, it will lead to 5-20% errors, which is a lot. Just I’m tired of arguing with my QA about the necessity of cross-comparison. It’s easier for them to save a few dollars an hour on the engineer's pay than to ensure proper error control, which is hardly avoidable without this check.

That's why I reached out to you for guidance in searching for specific statutes, administrative rules, and/or ordinances that regulate this scenario.

In the context of “physical roof boundary, roof boundary with gutters,” if you do UAV LiDAR scanning, UAV photogrammetry: for calculations related to solar panel installations, not only the total roof area but also its slope and shading load matter, which may change depending on whether gutters are included. If gutters are included in the roof perimeter, this could alter calculations for potential energy yield, shading analysis, and equipment placement.

Inaccuracies in this data may lead to suboptimal system designs. In this case, UAVs for photogrammetry can provide an overall view of the roof, while LiDAR-equipped UAVs can create precise 3D models that accurately account for gutters and slopes. (Energy Calculations and Solar Panel Installations) For various engineering and architectural calculations, it is necessary to know the area of the roof. The perimeter of the roof helps in the calculations by allowing you to determine exactly how much material will be needed to cover the roof and to estimate costs (Determining the roof area).

A couple of examples of projects in 2D, 3D

I am at a loss trying to figure out how anybody would possibly be using digital imagery from Google Maps or any other similar imagery providers.

It's nothing, it happens to everyone! 🙂 You are confused because Google Maps and Google Street View are different things.

Google Maps is a mapping service providing interactive maps, satellite imagery, navigation, and information about locations and routes. It allows users to plan routes for walking, driving, public transit, and biking. Google Maps also offers real-time traffic updates and recommends points of interest, restaurants, hotels, and many other places.

Google Street View is a feature within Google Maps that offers panoramic images of streets and locations. It allows users to "walk" through streets and view buildings, roads, and landmarks in a panoramic view. These images are collected using specialized cameras on vehicles, bikes, foot, and even drones for hard-to-reach places.

If you, for example, made an aerial survey and passed it for 2D and 3D planimetric mapping (decoding) to another person who was not physically present at the site, do you think he will have difficulties? 🙂 I worked for a total of 9 or 10 years, sorry I have lost count of the topographic engineer and always did field sketch. On very large projects such as topographic surveying town, we even had separate survey parties driving around the streets to record video and take note of the directions of wires, types of fences, utility elements (pipes, transformers) so that in case of a mistake there was a visual source of information to check.

A little advice: if you are doing an aerial survey and plan to pass it to another engineer for decoding, spend 10-15 minutes of your time and record video at the site. It may not cover the whole site, but in case of disputable issues during decoding aerial data, this video will make the work for the project executor very easy.
Thx

I use Google Street View when pricing a new job all the time. Gives a good perspective of the site. Best story; looking at a lot for a boundary survey and what shows up on the adjoining property, stakes and flagging from a recent survey! I was like, wow maybe we can make a buck on this one! LOL

"I am at a loss trying to figure out how anybody would possibly be using digital imagery from Google Maps or any other similar imagery providers."<div>

I don't think the OP is talking about mapping directly from Google et al. He is talking about using it a collateral information. It is no different than looking at site photos to clarify information collected in the field.

</div>

It's much different when it comes tracing digital imager, right down to gutter lines and showing them on a plan with offsets without understanding the coordinate systems involved, especially on an international level.

I have some experience of working in different coordinate systems, including the British (or Imperial) System of measurement; of course, I may not know something, we all learn new things from time to time.

For example, in Europe, in Belgium in particular, surveyors work in Belge 1972/ Belgian Lambert 72 (31370), but, including, they use the old coordinate system Belgian Lambert 2008 (3812). In Israel, it is Israeli Geodetic Datum 1995, IGD95, and all geodetic work is performed in Metric System of measurements.

In the U.S., it is, quote:
“Coordinate values should be based on National datums. Horizontal coordinate values should preferably be referenced to the North American Datum of 1983 (NAD 83). Vertical coordinate values should preferably be referenced to the North American Vertical Datum of 1988 (NAVD 88). However, it is recognized that many legacy maps and geospatial data are referenced to older national datums, such as the North American Datum of 1927 (NAD 27) and the National Geodetic Vertical Datum of 1929 (NGVD 29).”

If it is fundamental for you, Google Maps, OpenStreetMap, Bing Maps are based on EPSG:3857, Spherical Mercator. At one time, I even conducted a study for myself on how and with what accuracy qGIS performs transformation of Google Maps images from EPSG:3857 to EPSG:2272 - NAD83 / Pennsylvania South (ftUS).

If I understood you correctly and you meant road drainage, I am quite familiar with the United States Road Network. The “with offsets” thing I didn't really understand.

In the context of my original question, I quote, “Maybe there are specific statutes, administrative rules, and/or ordinances that regulate this scenario?” do you have any insights?