Et Logo vun OpenStreetMap OpenStreetMap

Kshitijraj Sharma's Diary

Recent diary entries

Exploring Overture Map Data

Posted by Kshitijraj Sharma on 9 June 2024 in English. Last updated on 14 June 2024.

Welcome ! First lets start with overture , if you don’t know what is Overture Maps Foundation and what it does I strongly recommend you to go through this website : https://overturemaps.org/ , I tried to build small utilities and hosted them so that readers of this blog also can look into the data and analyze by themselves.

Release Used

  • Overture release: 2024-05-16-beta.0

Objectives

Primary Objective

  • To perform qualitative and quantitative analysis of Overture map data.

Secondary Objectives

  • Visualize the releases on a country level.
  • Conduct qualitative analysis to identify additions to existing OSM data and differences across countries.
  • Facilitate general users in forming their own opinions based on the available data.

Approach

  1. Build a script to retrieve Overture data as geoparquet with multiple themes (streamlining and automating the process).
  2. Convert geoparquet to geojson.
  3. Convert flattened geojson to pmtiles.
  4. Develop a viewer for comparison and loading.
  5. Automate the entire process with a bash script.
  6. Compare with population data, existing OSM buildings in the area, and if possible, the number of people per building.

Considerations

See full entry

SAM and OSM

Posted by Kshitijraj Sharma on 8 June 2023 in English. Last updated on 15 June 2023.

Introduction

The purpose of this document is to explore the possibilities of the Segment Anything Model (SAM) Segment Anything developed by Meta in OpenStreetMap mapping. SAM is a promising model in the field of image segmentation, and its potential application to fAIr could enable the tracing of various features on maps. This document provides an overview of SAM’s working mechanism, test cases conducted with SAM, and two potential integration ideas for utilizing SAM

SAM Working Mechanism

SAM operates directly on images and generates individual masks based on the training datasets used during model training. Unlike object detection models, SAM focuses on segmenting all elements present in an image, regardless of their nature, based on texture, color, and points of difference from the background and several others can be found on their research doc SAM Working Mechanism

Advantages of SAM for fAIr and OSM Mapping

See full entry

Open StreetMap Bot on Twitter

Posted by Kshitijraj Sharma on 13 January 2023 in English.

Hi all ! I want to share the information about that I have created two OpenStreetMap Bot on Twitter

  1. Open Street Map Retweet Bot

https://twitter.com/OSM_retweet_bot

This Bot is created to follow the recent trends on OSM, What people are thinking or doing in OSM, This bot will try to retweet the tweets with hashtags : #osm, #openstreetmap and #hotosm, Runs every three hour

  1. Open Street Map Stats Bot

https://twitter.com/stats_osm

This Bot is an implementation of standalone open source tool called OSMSG https://github.com/kshitijrajsharma/OSMSG, This tool uses osmium to analyze the change files from OSM using different URLs and generates the stats in different file formats such as CSV, JSON, image etc. It has a GitHub action enabled which will run the stats for requested region and global and produces the stats in CSV and image format and shares it with the twitter community, You can follow the GitHub tool itself to see / download the stats. I have added Nepal as monitoring country, but if you want any other countries, you can request it via opening an issue or mentioning bot to twitter. https://github.com/kshitijrajsharma/OSMSG/tree/master/stats Since it uses GitHub, you can also go through the historical stats using GitHub history it will keep updating the same file when it runs and shares the link in twitter

Both projects are open source and beta release, I will be looking for the recommendations and improvements on future ! Thank you !