Mobile · AI/ML
HerbCam
Android app for herbal-plant recognition
Android application that classifies 30 herbal plant classes in real time or from gallery photos using a YOLOv5 model.
Overview
HerbCam is an Android application for identifying herbal plants. It detects and classifies plants either live through the camera or from an existing gallery photo, then presents the plant's description from an offline reference list.
Problem & solution
The problem
People who want to identify herbal plants in the field need an offline, camera-first tool rather than a search engine and a reference book.
What was built
A native Android app wrapping a YOLOv5 model for inference within the app, with two input paths — live camera and gallery — and a plant reference screen.
Key features
- Real-time plant detection through the device camera
- Classification from photos already in the gallery
- A 30-class herbal plant model
- Reference descriptions for each recognized plant
My contribution
- Android application development and model integration, as documented in the portfolio PDF.
Visual showcase
Related projects
View all
Mobile · AI/ML
TenunSantara
Telkom University — Bangkit Academy
Mobile app that recognizes Indonesian woven-cloth motifs from camera or gallery images and presents textile details.
Computer Vision · AI/ML
Smart Camera Lobster
Aruna Indonesia — R&D contract
Computer-vision system that enhances underwater footage, detects and counts lobsters, and tracks them across frames for remote monitoring.
Full Stack · Enterprise
Job Talentio
PT EGI Resources — IT project
Multilingual hiring platform connecting candidates, recruiters, and moderators — with CV import, weighted matching, and a recruiter pipeline.