Skip to content
IDPortal

Lecturer and student research groups

Geoinformatics Research Group

Reading the land from above.

We build artificial intelligence for geospatial data: deep learning and computer vision that detect, map and monitor crops, land and buildings from drone and satellite imagery, and deliver the results through geographic information systems and web and mobile applications.

Research map

Each cell is where one kind of media meets one verb. Choose a cell to see the theses and papers inside it.

DetectFind, count and assess objects in imagery: trees, crops and buildings.
MapTurn imagery into maps: land segmentation and location indexing.
DeliverTake models into the field through web maps and mobile applications.
Drone imagerySatellite imagery

Satellite imagery × Map

Turn imagery into maps: land segmentation and location indexing.

  • PaperWavelet-Enhanced State Space Model for Multi-Resolution Agricultural Segmentation in Satellite Time SeriesY. F. Hestrio, Y. I. Kurniawan, A. J. Mantau, W. Jatmiko, K. W. Veronica, D. S. Candra, A. K. Rizkita, M. H. Algifari, M. C. T. Manullang · IEEE International Conference on Aerospace Electronics and Remote Sensing Technology (ICARES) 2026, accepted · 2026
  • PaperSampling Strategy Outweighs Architectural and Loss-Based Approaches for Cloud Shadow Segmentation in Landsat ImageryA. K. Rizkita, Y. F. Hestrio, M. M. L. Ramadhan, M. H. Algifari, A. Febrianto, M. Verdiana, A. Iqbal · IEEE Geoscience and Remote Sensing Letters, under review · 2026
  • PaperMixed attention mechanism on ResNet-DeepLabV3+ for paddy field segmentationA. K. Rizkita, M. M. L. Ramadhan, Y. F. Hestrio, M. H. Hunafa, D. S. Candra, W. Jatmiko · TELKOMNIKA (Telecommunication Computing Electronics and Control), 23(6), 1611–1625 · 2025

Theses

This group offers thesis topics for undergraduate students.

Open thesis topics in this group:

  • Detecting and counting objects in drone imagery, such as coffee trees, maize plants and buildings.
  • Segmenting farmland and land cover in satellite imagery.
  • Assessing crop canopy health with deep learning.
  • Spatial indexing (geohash) and web-based monitoring maps.
  • Mobile applications that run computer vision models in the field.

Publications

6 journal and conference publications.

2026

  • Wavelet-Enhanced State Space Model for Multi-Resolution Agricultural Segmentation in Satellite Time SeriesY. F. Hestrio, Y. I. Kurniawan, A. J. Mantau, W. Jatmiko, K. W. Veronica, D. S. Candra, A. K. Rizkita, M. H. Algifari, M. C. T. Manullang · IEEE International Conference on Aerospace Electronics and Remote Sensing Technology (ICARES) 2026, accepted
  • Sampling Strategy Outweighs Architectural and Loss-Based Approaches for Cloud Shadow Segmentation in Landsat ImageryA. K. Rizkita, Y. F. Hestrio, M. M. L. Ramadhan, M. H. Algifari, A. Febrianto, M. Verdiana, A. Iqbal · IEEE Geoscience and Remote Sensing Letters, under review
  • Comparative Analysis of Deep Learning Architectures for Coffee Tree Detection from Aerial ImageryA. K. Rizkita, A. Febrianto, M. Verdiana, A. Iqbal, M. H. Algifari · Journal of Applied Informatics and Computing (JAIC), 10(2), 1383–1390

2025

2023

2022

  • Image-Based Building Count Detection Using Deep LearningR. Bagaskara, A. K. Rizkita, R. Fernandes, W. Yulita · Jurnal Sains Komputer & Informatika (J-SAKTI), 6(1), 94–100

Track record

  • 1Research grants
    • Coffee Tree Canopy Health Assessment using Deep Learning ApproachITERA Research Funding 2025, Research Group Strengthening scheme · 2025

Members