User:Blocal186/MapRurualPennsylvania

From OpenStreetMap Wiki
Jump to navigation Jump to search

This page is a proposal for #MapRurualPennsylvania MapRoulette challenges.

Discuss this proposal here: https://community.openstreetmap.org/t/proposal-for-mapruralpennsylvania-map-roulette-challenges/122673

Active Challenges

A challenge for Columbia County is currently live: https://maproulette.org/browse/challenges/50488

This challenge will be used for feedback, test engagement, and test quality of results.

Goal

Improve available information about roads in rural Pennsylvania counties. The end result would be roads that are tagged with the following tags

  • surface=*
  • lane_markings=*
  • lanes=*

Scope

The scope of this project is to improve road details in rural Pennsylvania. To do so, this project will focus on the lowest density areas of the state, Townships. Counties will be approached in reverse population order as listed on this Wikipedia article. The first ten are:

  • Cameron County
  • Sullivan County
  • Forest County
  • Fulton County
  • Potter County
  • Mountour County
  • Juniata County
  • Wyoming County
  • Elk County
  • Greene County


The scope of this challenge will be done one county at a time and filters for the following conditions.

highway =

  • secondary
  • tertiary
  • unclassified
  • residential
  • road

Out of Scope

The following are out of scope of this project.

  • service roads - Service roads connect businesses, buildings and houses to the main road system. Tagging all of these roads is not a priority for this project, and may be a task that can be re-visited at a later time.
  • highway=motorway, trunk, or primary - most of these roads are (a) complete, (b) complex or (c) require more validation than is appropriate for a short task.
  • Non-Townships - These areas are more densely populated and are easier worked on by area as apposed to by individual ways.

Challenges

Stage 1

Stage one is to use available imagery to determine if a road is `surface=paved` or `surface=unpaved`

Name: Identify COUNTY_NAME_HERE, Pennsylvania road surfaces.

The Overpass query targeting `ways` is as follows:

//!!! Remove the below line when using on MapRoulette !!!
[out:xml][timeout:25];

// Limit the search area to a single county
area['border_type'='state']['name'='Pennsylvania']->.state;
area['name'='COUNTY County'](area.state)->.county;

// Exclude high-population areas (non-townships)
rel
  (area.state)
  (area.county)
  [admin_level=8]
  [border_type!=township];

map_to_area -> .exclusion_area;

// Search for ways
way
 ["highway"~"(secondary|tertiary|unclassified|residential|road)(_link)?"]
 [!surface]
  (area.state)
  (area.county)
  ->.a;

// Search for ways in the exclusion area
way.a
  (area.exclusion_area)
  ->.b;

// Subtract set B from A
( .a;- .b;);

// Remove the below line when using on MapRoulette
(._;>;); 

// Output the results
out geom;

Description

 Help identify rural road surfaces. Use available imagery to identify if a road should be tagged with `surface=paved` or `surface=unpaved`. 


How should your Challenge be Categorized?

 Roads / Pedestrian / Cycleways


Detailed Instructions for Mappers

 Use available imagery to identify if a road should be tagged with `surface=paved` _or_ `surface=unpaved`. Exact identification is not necessary. Roads concealed by trees should be marked as *Can't Complete* with an MR tag of `bad-imagery`. 
 Please note: Imagery alignment in this area may not be accurate, do not modify geometries. However if you encounter a road that seems to just not exist, please mark the task as *Can't Complete* with an MR tag of `ghost-road`.
 Tips: 
 * Roads with lane markings are likely paved. However, roads without markings are not automatically unpaved. 
 * Roads with wheel tracks are likely unpaved.
 * Unpaved roads tend to leave gravel trails onto connecting paved roads.

Difficulty of your task

 Easy


Changeset Description

 Identified road surface. #MapRuralPennsylvania  #maproulette

MR Tags

  • ghost-road: Roads that don't appear to exist. Probably bad tiger import data.
  • bad-imagery: Unable to complete the task with given imagery

MR Review Tags

  • TBD

Request review by default

 Yes

Stage 2

Identify if a road has lane markings. Ignore roads that already have the number of lanes marked, as this is the precursor to the next step.

New Way search query

// Search for ways
way
 ["highway"~"(secondary|tertiary|unclassified|residential|road)(_link)?"]
 [surface~"(paved|cement|asphalt)"]
  [!lane_markings]
  [!lanes]
  (area.state)
  (area.county)
  ->.a;

Stage 3

Identify number of lanes by lane markings

New Way search query

// Search for ways
way
 ["highway"~"(secondary|tertiary|unclassified|residential|road)(_link)?"]
 [surface~"(paved|cement|asphalt)"]
  [lane_markings=yes]
  [!lanes]
  (area.state)
  (area.county)
  ->.a;