Dataset Viewer
The dataset viewer is not available for this dataset.
Unexpected token '<', "<html> <h"... is not valid JSON

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Bakkhali River Estuary — High Tide Boat Survey (Free Sample)

50 GPS-tagged coastal images from a single high-tide boat survey of the Bakkhali River estuary, Khurushkul, Cox's Bazar, Bangladesh. By Golam Rob — www.golamrob.com

Free to use, including commercially — just credit "Golam Rob (golamrob.com)". Licensed CC BY 4.0. Use it, train on it, remix it, share it. All I ask is attribution.

📸 These 50 images are a small taste of a 200,000–300,000 image personal library of coastal, tidal, and mangrove imagery from Bangladesh. If you need more — specific locations, tide states, volumes, or labeled/segmented data — reach out: www.golamrob.com · hello@golamrob.com


Dataset Summary

These 50 photographs were captured within a single 74-minute window on 4 July 2026 (11:17–12:31, UTC+6) from a traditional wooden boat navigating the Bakkhali River estuary at peak high tide, while Cox's Bazar port had Bangladesh Meteorological Department Local Warning Signal No. 3 hoisted (stormy/squally conditions over the North Bay of Bengal).

The low, on-water viewpoint captures the waterline between the estuary and its mangrove fringe from an angle not obtainable from satellite or aerial platforms. Useful for:

  • Waterline / tidal-inundation segmentation
  • Coastal flood-risk and storm-surge model grounding
  • Mangrove and intertidal-zone monitoring
  • Pretraining / fine-tuning vision models on coastal Bangladesh scenes

Supported Tasks

  • Image segmentation — separating water, waterline, and mangrove vegetation (segmentation masks are not included in this free sample).
  • Image classification / scene understanding — coastal, high-tide, estuarine scenes.

Languages

All metadata (titles, captions, keywords) is in English.

Dataset Structure

Loads as a single split of 50 images with per-image metadata (Hugging Face imagefolder format). Each image is a preview-resolution JPG (2000 × 1333, ~2.7 MP), a downscaled 1/3 export of the original 24 MP NEF (RAW), with no embedded EXIFmetadata.jsonl is the single source of truth.

Data Fields

Field Type Description
file_name string Image filename (links row to image)
sha256 string SHA-256 checksum of the delivered image
title string Series title
caption string Scene / dataset description
date_captured_iso string Capture time, ISO 8601, UTC+6
gps_latitude, gps_longitude float Fuzzed to ~100 m (see below)
gps_precision string Fuzzing note
gps_altitude string Raw GPS-derived, uncorrected (see Notes)
camera_make, camera_model, lens string Nikon D7200; 50 mm f/1.8 or 18–140 mm
focal_length, exposure_time, f_number, iso string Capture settings
width_px, height_px, megapixels number 2000 × 1333, 2.7
keywords string Semicolon-separated tags
identifiable_people_visible string People/consent note
city, region, country string Cox's Bazar, Chattogram, Bangladesh
copyright, credit string Golam Rob
post_processing, color_space, source_format string Provenance notes

Dataset Creation

  • Curated and photographed by: Golam Rob (professional photographer / data curator), from a personal library of 200k–300k coastal & tidal images of Bangladesh.
  • Equipment: Nikon D7200 (24 MP DSLR); Nikkor 50 mm f/1.8 and 18–140 mm lenses.
  • Method: continuous estuarine transect from a hired wooden boat at peak high tide; GPS track logged separately by a phone GNSS app; images time-synced to the track.
  • Processing: direct NEF → JPG export via Nikon NX Studio 1.10, downscaled to 2000 × 1333, with no exposure/colour/contrast edits; delivered as a metadata-stripped preview JPG.

Notes on the Data

  • GPS coordinates are fuzzed to ~100 m (random jitter) in this sample. This is a privacy choice, not a quality limit — it does not affect visual use of the images.
  • Embedded EXIF has been stripped from all images (no serial number, no raw GPS).
  • Altitude is raw GPS-derived and uncorrected (typical vertical GPS error ±10–20 m). This was an on-water survey, so per-image altitude is approximate, not surveyed elevation.
  • Identifiable people: all 50 sample images were reviewed and show no identifiable people.
  • Scope: single location, single day, high tide only — a small sample, not a full training corpus. For broader coverage (locations, tide states, volume, labels), see www.golamrob.com.

Licensing Information

CC BY 4.0 — free to use, share, and adapt for any purpose, including commercially, provided you give appropriate credit.

Citation

@misc{golamrob_bakkhali_hightide_sample_2026,
  title  = {Bakkhali River Estuary — High Tide Boat Survey (Free Sample)},
  author = {Golam Rob},
  year   = {2026},
  howpublished = {Hugging Face Datasets},
  note   = {CC BY 4.0. https://www.golamrob.com}
}

Contact / More Data

Golam Rob — coastal & tidal imagery datasets (200k–300k image library) 🌐 www.golamrob.com · ✉️ hello@golamrob.com

Please credit Golam Rob (golamrob.com) when you use these images. Thank you!

Downloads last month
100