Malaysia Rainforest & Wildlife Data

Simple reports.
Smarter protection for Malaysia’s wildlife.

GeoWild connects people across Malaysia with conservation teams. Report wildlife sightings, roadkill and human–wildlife conflicts with GPS tags – your data supports hotspot alerts, ecological research, and long-term planning.

Photos that you share also help train machine learning models that learn to recognise Malaysia’s species – from large mammals and birds to reptiles, fish, insects, flora and aquatic life.

The platform is designed for the Malaysian context: tropical rainforests, rivers, coastal zones, plantations and cities – all in a single, unified biodiversity reporting system.

Privacy-first: Your data is sent securely to relevant authorities only. It is not sold, rented, or shared with private companies.

About GeoWild

A national bridge between citizens, rainforest data, and wildlife authorities

GeoWild channels reports from the public – from villages, highways, plantations and urban parks – into a structured biodiversity database to support field response, machine learning models and scientific analysis.

Biodiversity & Habitat Mapping

Wildlife and plant sightings with precise coordinates help build species distribution maps across Malaysia’s rainforests, rivers, coasts and cities.

Roadkill Hotspot Detection

Roadkill reports – including species, time and direction of travel – support identification of road hotspots, future warning alerts, and potential wildlife corridors.

Verified, Secure & AI-ready Data

Each report carries location, time and category meta-data. Records are quality-checked and stored securely, then used to support dashboards, research and training of machine learning models that help suggest possible species from images.

Technology in Conservation

Built on maps, sensors and machine learning

GeoWild uses modern web technologies, GPS, mapping services and machine learning to transform public reports into actionable conservation intelligence. Patterns of roadkill, conflict and species presence can be visualised on dashboards used by authorities.

  • GPS-locked locations enable accurate field response and spatial analysis.
  • Map context combines reports with habitat, land-use and road networks.
  • Machine learning image models learn from citizen photos to suggest possible species, from large mammals to birds, reptiles, fish, insects, trees and aquatic plants.
  • Secure APIs connect GeoWild to internal authority systems for official use.

Data & Image Flow Overview

  1. 1 You submit a report of wildlife, roadkill or conflict with optional photo and description.
  2. 2 GeoWild validates and protects your data, stores it securely and flags images for machine learning training where appropriate.
  3. 3 Authorities receive and analyse records for field response, monitoring and policy planning, while image datasets help improve automated species recognition for Malaysia.

GeoWild is an independent platform. It does not replace official e-complaint portals, but passes on citizen reports in a structured format to help public agencies.

AI Species Helper

How our AI learns from your images

GeoWild uses TensorFlow.js in the browser to run lightweight image models that suggest possible species from your photos – without sending raw images to third-party AI services. Over time, anonymised images from citizens help the model learn Malaysia’s biodiversity better, from big mammals like elephants and tapirs to common birds, reptiles, fish, insects, trees and algae.

Step 1

You upload a clear photo

A simple phone photo of the animal, plant or roadkill, with GPS and basic notes, is enough to help improve the model.

Step 2

TensorFlow.js runs in your browser

The image is resized and passed through a trained model. The AI produces scores for a list of non-sensitive species classes such as elephant, tapir, macaque, hornbill, monitor lizard and others.

Step 3

You see a suggestion – not a final verdict

The result is shown as a suggestion only. Final identification remains with trained officers and researchers. The goal is to speed up triage, not to replace experts.

Note: The on-page demo uses a placeholder model path (/models/geowild-species-tfjs/model.json). Deployers should replace this with a trained TensorFlow.js model and ensure class labels match the code.

Malaysia Map (Demo)

Sample roadkill and biodiversity reports

This section shows a demonstration of how GeoWild could visualise reports across Malaysia. The locations and records below are examples only, used to illustrate how roadkill and wildlife sightings might appear on a national map.

Map, markers and counts are non-real demo data for interface preview.

Total reports (demo)

1,284

Citizen reports submitted across Malaysia.

Animals recorded (demo)

3,452

Unique animal records in the GeoWild database.

Roadkill reports (demo)

217

Recorded roadkill cases for hotspot analysis.

Malaysia – Demo Hotspots
Sample records (non-real)
RK

Roadkill – Tapir (demo)

Malayan Tapir – East–West Highway (Gerik area)

Reported by driver at night; adult tapir found on road shoulder in a forested stretch. Used in demo to illustrate potential alert for frequent collisions.

Category: Roadkill · Habitat: Forest edge / highway

RK

Roadkill – Elephant (demo)

Asian Elephant – trunk road near plantation

Demo record showing how large-mammal roadkill could trigger long-term monitoring, signage or speed calming measures in collaboration with authorities.

Category: Roadkill · Habitat: Plantation / forest mosaic

OB

Observation – Macaque (demo)

Long-tailed Macaques near village fruit trees

Group of macaques feeding on rambutan trees at village edge. These kinds of reports help map human–wildlife interaction zones for awareness and mitigation.

Category: Mammal · Interaction: Non-aggressive presence

OB

Observation – Hornbill (demo)

Large hornbill flying over forested valley

Example of how bird observations from hikers and local communities can fill data gaps on forest connectivity and fruiting tree availability.

Category: Bird · Habitat: Forested valley / ridgeline

Wildlife Photo Album (Demo)

Examples of langurs, macaques and birds

A visual album of Malaysian wildlife – including dusky leaf monkeys (lotong), long-tailed macaques (kera) and rainforest birds. In the real system, similar albums can be generated automatically from verified citizen reports and AI-tagged images.

All photos in this section are royalty-free demo images from Unsplash, used only for illustration. They are not real GeoWild reports.

Browse album

These examples show the type of citizen-uploaded images that can train GeoWild’s models to recognise species groups such as langurs (lotong), macaques (kera) and common forest birds across Malaysia.

Species Explorer (Demo)

Examples of Malaysian biodiversity

Below are sample species from several groups commonly found in Malaysia. Click a card to view a sample image and basic information. In a full system, these records would be driven by real citizen observations and machine learning suggestions.

All demo photos in this section use royalty-free images (e.g. Unsplash) and are not taken from real GeoWild reports.

Malaysia Biodiversity Records (Demo)

Public Observations & Roadkill Reports

A preview of how GeoWild can display wildlife sightings, flora, fauna and roadkill reports submitted by the public. Records shown are demo-only.

Malaysia Demo Hotspots

Dusky Leaf Monkey (Trachypithecus obscurus)

Teluk Bahang, Penang

Today · 09:15

Long-tailed Macaque (Macaca fascicularis)

Kuala Selangor

1 h ago

Roadkill – Tapir (demo)

Gerik Highway

2 days ago

Trusted References & Authorities

Who has legal authority – and how GeoWild fits in

The agencies below are the official authorities with legal powers to investigate, enforce and take action on wildlife, forestry and emergency matters in Malaysia. GeoWild is not part of these agencies. The platform only helps citizens submit structured information which is then shared with them on a voluntary, public-interest basis.

Global Conservation Reference

IUCN logo

IUCN Red List of Threatened Species

Global database providing conservation status and categories for species worldwide, used as an international reference.

Visit iucnredlist.org

Malaysia Wildlife, Forestry & Emergency Authorities

PERHILITAN logo

Department of Wildlife and National Parks (PERHILITAN)

Federal authority for wildlife protection, licensing and protected areas in Peninsular Malaysia.

Visit wildlife.gov.my
Forestry Department Peninsular Malaysia logo

Forestry Department of Peninsular Malaysia (JPSM)

Manages, plans and protects Permanent Forest Reserves and forestry resources in Peninsular Malaysia.

Visit forestry.gov.my
Sarawak Forestry Corporation logo

Sarawak Forestry Corporation (SFC)

Statutory body managing Totally Protected Areas and biodiversity conservation in Sarawak.

Visit sarawakforestry.com
Sabah Wildlife Department logo

Sabah Wildlife Department (Jabatan Hidupan Liar Sabah)

State authority responsible for wildlife conservation and enforcement in Sabah under the Wildlife Conservation Enactment 1997.

Visit wildlife.sabah.gov.my
APM logo

Malaysian Civil Defence Force (APM)

Supports disaster and emergency response, including selected wildlife-related incidents during floods and other events.

Visit civildefence.gov.my
JBPM logo

Fire and Rescue Department of Malaysia (JBPM)

National fire and rescue agency handling emergency response and rescue, including certain animal rescue cases.

Visit bomba.gov.my

GeoWild operates independently and does not represent or speak on behalf of any of the agencies above. Only reports voluntarily submitted by the public are shared with them, to support official work for the benefit of people and wildlife.

Safety & Privacy

How GeoWild protects your data

GeoWild is designed for public use, but your personal details and exact locations are treated as sensitive information.

  • We collect only the information needed to understand your report and support official response.
  • Data is transmitted over secure connections to internal authority systems.
  • GeoWild is not a public feed or social network – your report is not displayed on a public map.

Data & image sharing commitment

  • Your data will only be shared with relevant government agencies and recognized conservation bodies.
  • We will not sell, rent or provide your personal data to third parties for marketing purposes.
  • Photos may be retained in anonymised form to train GeoWild’s machine learning models so that the system can better suggest likely species from future images.
  • You may request clarification on how your data is stored and used via official help channels.

Using GeoWild does not replace emergency hotlines. In life-threatening situations, always contact local emergency services first.