# Dimas Maulana Ichsan (dmxsan) | Full Technical Corpus > Portfolio of Dimas Maulana Ichsan (dmxsan): applied remote sensing, GIS engineering, and GeoAI pipelines delivered as interactive spatial plates. ## Author Identity & Telemetry - Name: Dimas Maulana Ichsan - Handle: dmxsan - Domain Focus: Remote Sensing, GIS & GeoAI - Base Location: Yogyakarta, ID (YOG // 7.7956° S, 110.3695° E) - Canonical URL: https://dmxsan.dev/ - Index Summary: https://dmxsan.dev/llms.txt - RSS Feed: https://dmxsan.dev/rss.xml - Core Competencies: Remote Sensing, Earth Observation, Geographic Information Systems (GIS), GeoAI, Sentinel-1 SAR, Flood Mapping, Deep Learning, Spatial Engineering ## Background & Engineering Philosophy Dimas Maulana Ichsan (dmxsan) is a remote sensing and GIS specialist working across earth observation and GeoAI. His work applies multi-sensor satellite data to study terrestrial ecosystems, land-cover dynamics, and long-term environmental change. With an emphasis on sound engineering practice, the focus goes beyond one-off analysis to complete workflows: reproducible earth observation pipelines, clean spatial modeling, and interactive interfaces built for practical inspection. ### Technical Focus Areas - **Remote Sensing & GeoAI:** Multi-sensor satellite imagery analysis, machine learning workflows, and large-scale land-cover classification - **GIS & Environmental Monitoring:** Spatial modeling, forest and vegetation dynamics, ecosystem assessment, and earth-change analysis - **Geospatial Engineering:** Interactive spatial plates, map tile workflows, and high-performance geospatial web systems ## Profiles & Verified Channels - GitHub: https://github.com/dmxsan - LinkedIn: https://www.linkedin.com/in/dimasmaulana99/ - Email: mailto:hello@dmxsan.dev - Kaggle: https://www.kaggle.com/dimskeee --- ## Technical Projects & Case Studies ### Mapping the Nakba: Historical WebGIS & Cartographic Archive - URL: https://dmxsan.dev/projects/mapping-the-nakba/ - Category: WebGIS · Historical Cartography - Domain: WebGIS - Status: active - Published: 2026-05-28 - Stack Tags: MapLibre GL, Vue 3, TypeScript - Source Repository: https://github.com/dmxsan/mapping-the-nakba - Live Application: https://mapping-the-nakba.vercel.app **Summary:** Interactive MapLibre GL and Vue 3 cartographic archive layering 1875–1951 historical Palestine Open Maps surveys over present-day satellite imagery with an event timeline. #### Specifications - **Map Engine:** MapLibre GL JS (multi-layer raster & GeoJSON) - **Frontend Stack:** Vue 3, TypeScript, Pinia, Tailwind CSS - **Historical Layers:** 6 Palestine Open Maps tilesets (1875–1951) - **Localization & Search:** Bilingual EN / AR with Fuse.js ## Overview **Mapping the Nakba** is an open-source WebGIS application built to examine territorial change across Mandatory Palestine, the West Bank, and the Gaza Strip from 1947 to the present. Instead of static side-by-side images, the viewer lets users blend archival survey maps directly over modern **Esri World Imagery** to compare historical village footprints against current land cover. ## Cartographic & Interactive Features - **Historical Survey Overlays:** Integrates 6 archival tile layers from **Palestine Open Maps** spanning 1875–1951, including British Survey of Palestine sheets, RAF aerial photography, and early survey editions with per-layer visibility and opacity controls (`LayerControl.vue`). - **Temporal Event Timeline:** Tracks 11 key historical periods from 1948 through 2023–2026 (`TimelineSlider.vue`), dynamically highlighting affected administrative and regional GeoJSON polygons (`mandatory-palestine`, `west-bank`, `gaza`, `jerusalem-area`, and neighboring districts) in `WebGISViewer.vue`. - **Village Dataset & Search:** Pairs `villages.geojson` and regional geometry overlays with client-side fuzzy search powered by **Fuse.js** inside a **Pinia** state store. - **Bilingual Support:** Full English and Arabic interface support across event descriptions, layer controls, and map metadata. ## Engineering & Deployment Notes Built with **Vue 3** (Composition API), **TypeScript**, and **Vite**, keeping all GeoJSON overlays, timeline state, and raster tile blending lightweight in the browser and deployed continuously on **Vercel**. ### Deep Learning U-Net for Radar Imagery Flood Mapping - URL: https://dmxsan.dev/projects/radar-flood-unet/ - Category: GeoAI · Flood Mapping - Domain: GeoAI - Status: active - Published: 2021-09-07 - Stack Tags: SAR, U-Net, TensorFlow, Flood Mapping - Source Repository: https://github.com/dmxsan/Deep-Learning-using-UNet-for-Radar-Imagery-Flood-Mapping - Dataset: https://www.kaggle.com/dimskeee/unet-segmentation-using-radar-imagery - Interactive Atlas Plate: https://dmxsan.dev/atlas/radar-flood-unet/ **Summary:** Semantic segmentation of Sentinel-1 polarimetric SAR imagery using a TensorFlow U-Net trained on 128×128 radar chips for rapid post-flood mapping. #### Specifications - **Architecture:** U-Net (16 / 32 / 48 / 64 filters) - **Sensor Input:** Sentinel-1 Polarimetric SAR (128×128 chips) - **Training Stack:** TensorFlow / Keras · RTX 2060 Super (6 GB) - **Research Origin:** PKM-RE · PIMNAS 34 (UGM, 2021) ## Problem Context During heavy rainfall and floods, cloud cover often blocks optical satellites. **Sentinel-1** Synthetic Aperture Radar (SAR) penetrates clouds and responds to surface roughness and scattering behavior. Because calm floodwater acts like a specular reflector, it scatters radar energy away from the sensor and appears sharply distinct from surrounding vegetation and built-up land in polarimetric composites. ## Research Origin This pipeline originated from a 2021 **PKM-RE**[^1] research project at the **Faculty of Geography, Universitas Gadjah Mada (UGM)** (*Analisis Pemetaan Cepat Pasca Bencana Banjir Menggunakan Dekomposisi Polarimetrik Radar Sentinel-1 Berbasis Model Deep Learning*), representing UGM at **PIMNAS 34**[^2]. - **Team:** Iklila Rahmatika (Lead), Elyana Inggrid Widiastuti, Dimas Maulana Ichsan, and Made Dwiyana Surya Upadana - **Advisor:** Dr. Sc. Sanjiwana Arjasakusuma, S.Si., M.GIS. ## Model & Chip Pipeline Sentinel-1 scenes (including South Kalimantan flood scenes around `2.38°–2.85° S, 115.07°–115.15° E`) are prepared as **128 × 128 pixel** polarimetric false-color chips paired with binary water masks and published on Kaggle (`dimskeee/unet-segmentation-using-radar-imagery`). - **Encoder-Decoder Network:** A compact U-Net using `16`, `32`, `48`, and `64` filter blocks with skip connections to preserve narrow river channels and flooded boundaries. - **Memory-Friendly Training:** Patch size and filter counts stay light enough to train smoothly on a **6 GB VRAM** GPU (RTX 2060 Super) without losing boundary precision. - **Visual Output Check:** Stitched `512×512` cosine-blended predictions (`IoU 0.74–0.83`) can be inspected pixel-by-pixel against SAR backscatter and ground truth in [Plate 01 (/atlas/radar-flood-unet/)](/atlas/radar-flood-unet/). ## Engineering Notes Keeping preprocessing and inference lightweight makes it straightforward to export predicted flood masks into standard GIS deliverables, such as GeoTIFF rasters or vector polygons, for rapid post-disaster response workflows. [^1]: **PKM-RE** (*Program Kreativitas Mahasiswa - Riset Eksakta*): Indonesia's national competitive undergraduate exact-sciences research grant program funded by the Ministry of Education, Culture, Research, and Technology (Kemendikbudristek). [^2]: **PIMNAS 34** (*Pekan Ilmiah Mahasiswa Nasional ke-34*): The 34th National Student Scientific Week (2021), Indonesia's annual national finals where top PKM-funded university research teams present and compete. --- ## Interactive Spatial Atlas Plates ### Plate 01: Sentinel-1 SAR Flood Prediction - URL: https://dmxsan.dev/atlas/radar-flood-unet/ - Domain: Flood Mapping - Viewer Engine: chip-inspector - Sensor: Sentinel-1 SAR (C-Band) - Resolution: 512 × 512 mosaic (4×4 × 128px) - Dataset: Kaggle Radar Flood Chips - Spec Badge: 512×512 SAR · U-Net - Published: 2024-05-01 - Related Project Case Study: https://dmxsan.dev/projects/radar-flood-unet/ **Summary:** Interactive 512×512 mosaiced Sentinel-1 SAR viewer comparing backscatter amplitude against U-Net flood masks, ground truth, and shoreline error overlays. #### Inspected Raster Samples - **Chip 01 · Main River Channel** (2.38° S, 115.07° E): Predicted Extent 34.1% | GT Extent: 36.8% | IoU: 0.74 - **Chip 02 · Broad Floodplain** (2.85° S, 115.15° E): Predicted Extent 50.1% | GT Extent: 51.9% | IoU: 0.83 - **Chip 03 · Meandering Channel & Floodplain** (2.40° S, 115.07° E): Predicted Extent 37.7% | GT Extent: 40.5% | IoU: 0.77 ## Plate Interpretation Each **Plate** in `~/atlas` pairs a geospatial model output with a viewer built for that data type. This plate inspects contiguous `4×4` (`512×512` px) Sentinel-1 scenes reconstructed using overlapping `stride=32` cosine-blended inference to eliminate `128×128` tile seams: - **SAR False-Color RGB Base:** In the 3-band Sentinel-1 input composite, purple/magenta tones indicate smooth flooded or open-water surfaces while green tones indicate rough or dry land cover. - **Overlay Mode:** Blends the selected binary mask over the Sentinel-1 false-color base, marking water perimeter boundaries in NIR crimson. - **Split Mode:** Slides between raw SAR False-Color RGB on the left and the predicted mask, ground truth, or agreement diff on the right. - **Mask Mode:** Isolates the binary classification or pixel-wise agreement map on a clean canvas for shape and boundary inspection. - **Layer Switch:** Compare **U-Net Pred** directly against **Ground Truth** or inspect **Pred vs GT Diff** (True Positive, False Positive, and False Negative pixels).