ZryDys

Geospatial & Imagery analysis with open source

We aim at increasing safety by map analysis for rescue

What: Simple GIS and Open Source tools to analyze locations.

How: transparently share open tools and offer support, ranging from business projects to sea rescue. Circular reusing of COTS or existing devices.

Why: Conscious about safety in rescue and finding missions.

Where: Testing at ElHierro & Ondarroa, more sites under consideration

Status: TRL4 x CRL3 . Lab demo of integrated system x Validate the Market hypotheses

Affordable Overall the project relies on open sources and existing devices and transparent collaboration to disseminate know-how and be able to deliver on-time and on-budget:

Mission: to bridge the tech gap using open-source tools, training, and low-cost innovation — so even small projects benefit from GIS analysis once reserved for organizations* **.

Examples of affordable solutions: adding demo or redundancy:

  • Our online demo to interact and display multiple, geospatial views layers for electronic light table https://zrydys.github.io/elt.html

Promoting circular economy solutions where we can share, reuse, repair, refurbish existing devices. (contributing also to diminish pollution on disposal of old cellular or laptops)

ONGOING JULY AUG 2025 … want to join or collaborate?: 📞 Contact us: | ✉️ ZryDys@gmail | 🌐 ZryDys.github.IO

– refs gradually added below: #map #gis #raspb #pi

🌱 Practical Solutions

An open and agile mindset delivers smart, reusable solutions in today’s circular economy. This GIS summary is freely shared — and we’re here to support you with implementation, training, and guidance every step of the way.

References of powerfull GIS solutions that our customers can use, profit of our experience with: NOAA AIS Data (ESRI https://marinecadastre.gov/accessais/ https://marinecadastre.gov/accessais/ https://coast.noaa.gov/arcgis/rest/services/MarineCadastre/AISVesselTransitCounts2024/MapServer https://github.com/ocm-marinecadastre/ais-vessel-traffic )

Technical GIS tools

Satellite imagery analysis, processing software and techniques as capabilities. Electronic Light Table to view and manipulate images in real-time, or add graphics and text on top.

QGIS supports a variety of open source data formats and plugins to explore and integrate geospatial data from diverse sources. It can act as an ELT in the sense of managing and analyzing large datasets of satellite or aerial imagery.

GRASS GIS work with time series data for geospatial data management and analysis, image processing, graphics and map production.

Opticks open source for electronic light table (ELT), alike commercial ERDAS IMAGINE GLT, RemoteView, ENVI, or SOCET GXP WebView, we can add extensions similar to other remote sensing tool. Used at NASIC for for non-literal exploitation https://github.com/opticks-org.

Whitebox GAT visualize data in 3D, used in the environmental field and academic.

gVSIG has various data formats (Shapefiles, GeoTIFFs…) and the geospatial data management includes vector and raster data analysis, network analysis, and 3D.

Hubs

  • ESA Sentinel Hub cloud-based platform for visualizing and analyzing satellite imagery. there is a free tier available for light usage. Real-time satellite imagery from ESA satellites

  • NASA WorldWind open-source virtual globe to interact with global satellite imagery. The platform supports custom imagery layers, vector maps, and geospatial data overlays. It can be useful for building custom ELT tools.

Servers

  • MapServer open-source platform for publishing spatial data and interactive mapping applications. raster and vector data

  • GeoServer open-source server for sharing, display, edit, and process geospatial data with standard web services (WMS, WFS, WCS).

  • TileStache Python-based server for generating and serving raster tiles, useful for map-based imagery applications. useful for ELT processes.

Coding

  • OpenLayers & Leaflet open-source JavaScript library for displaying maps and geospatial data, interactive web-based ELT applications, to work with imagery in the browser.

  • GDAL low-level, open-source library for reading and writing raster and vector geospatial data formats or scripting to process ELT-style workflows.

  • Examples of Imagery Analysis ELT applications coded in C++ https://techpubs.jurassic.nl/library/manuals/1000/007-1387-050/sgi_html/apg.html

Formats NITF 2.0/2.1, GeoTIFF, ENVI, ASPAM/PAR, CGM, DTED, Generic RAW, ESRI Shapefile, HDF5, AVI, MPEG, JPEG, GIF, PNG, BMP

You can visualize the folloiwng charts in github source documents: https://github.com/zrydys/zrydys.github.io/blob/master/gis.md

graph TD;
    sat_imagery-->process;
    sat_imagery-->proC;
    process-->analysis;
    proC-->analysis;

..

sequenceDiagram
   autonumber
   imagery->>Proc: send to process
   loop checks
       Proc->>Proc: processing tools
   end
   Note right of Proc: open sources!
   Proc-->>imagery: ack!
   Proc->>analysis: several analysis
   analysis-->>Proc: feedback!

🥇 QGIS GUI Linux practical alignment:

Sucess cases: We align (georeference) mapping images:

  • qGIS transparent overlay let us visually match borders
  • see our clip/export tiles in a 6-piece map demo WMS

Example Workflow

  1. Open QGIS
  2. Load your 6-tile political map
    • If they’re separate PNGs, load them all (or merge them first)
  3. Load the satellite image to align
  4. Use Georeferencer tool:
    • Pick common points (coastlines, borders, corners)
    • Match satellite map → political map
    • Use simple transform (Affine or Polynomial 1)
  5. Once aligned:
    • Use Raster → Extraction → Clip Raster by Extent
    • Define the same extents as your 6 tiles
    • Export into 6 matching PNGs

👉 This gives you perfect overlap (same pixel grid, same bounds)

🥈 Semi-automatic coding approach Jupyter

Scripting this using OpenCV (feature matching) and rasterio (cropping/tiling)

See basic example which: Detects matching features between maps (edges, coastlines), Compute transformation (homography), Warp satellite image, …

import cv2  
import numpy as np  
  
img1 = cv2.imread("political.png", 0)  
img2 = cv2.imread("satellite.png", 0)  
  
orb = cv2.ORB_create()  
kp1, des1 = orb.detectAndCompute(img1, None)  
kp2, des2 = orb.detectAndCompute(img2, None)  
  
bf = cv2.BFMatcher(cv2.NORM_HAMMING)  
matches = bf.match(des1, des2)  
  
#% cj compute homography  
src_pts = np.float32([kp1[m.queryIdx].pt for m in matches]).reshape(-1,1,2)  
dst_pts = np.float32([kp2[m.trainIdx].pt for m in matches]).reshape(-1,1,2)  
  
H, _ = cv2.findHomography(dst_pts, src_pts, cv2.RANSAC)  
aligned = cv2.warpPerspective(img2, H, (img1.shape[1], img1.shape[0]))

above required in RHEL type linux:

  • sudo dnf install python3-opencv

Regarding AI & openCv: OpenCV can Load neural networks (DNN module) and Run models from TensorFlow / PyTorch, So:

  • Using ORB / Canny / homography → classical OpenCv ❌ not AI
  • Running a neural net through OpenCV → related to ✅ AI/ML

| Concept | What it is | Example (Linux / Python / Web) | | ——- | ———————————————- | ——————————- | | AI | Broad goal: machines doing “intelligent” tasks | Chatbots, planning systems | | ML | Subset of AI: systems learn from data | Training a classifier in Python | | DL | Subset of ML: neural networks | PyTorch CNN detecting objects |

Our charts in this page

  • Quick & portable → Mermaid
  • optional: Precise shapes (triangles etc.) → SVG
  • Future-looking → D2

Refs

References include

🌐languages:
DE🇦🇹🇩🇪 FR🇨🇦🇧🇪 NL🇳🇱🇧🇪 ES🇪🇸🇨🇷 EN🇺🇸🇮🇪