Ichthyological Society of Singapore launches free SGFish app for fish identification

On September 17, 2026, the Ichthyological Society of Singapore launched a free mobile application called SGFish on the Apple App Store, as reported by The Straits Times. The application uses an image recognition model to identify more than 900 fish species found in Singapore’s murky waters. Users snap a photograph of their catch through the app, which then identifies the species and logs the location of the find. To maintain scientific integrity, volunteers and co-founders of the society manually verify all submitted records before they enter an official database.

Building the Computer Vision Model and Database

Brian Sng, a 21-year-old software engineering student at Singapore Management University and lead developer of the mobile app, collated and vetted more than 90,000 photos of fish species from international databases, society members, and his own archives. It took two months to train the computer vision model until it reached an 82.7% accuracy rate. Sng noted that this model is more accurate than the model created by the US-based flora and fauna database iNaturalist, which maintains an accuracy rate of about 80% for fish globally. Before the app launched, the database was compiled using a Telegram bot called Angler’s Pokedex, built by 36-year-old cybersecurity professional and volunteer Samuel Pua. Pua explained that the team developed the app because time and location details could not be reliably extracted from images submitted on Telegram. Currently, the database contains about 2,300 observations covering more than 400 species.

App Alerts Users to Release Endangered Fish Species

The development of the app was spurred partly by recurring incidents of misidentification among recreational fishers. In August at Bedok Jetty, an angler mistook a globally endangered zonetail butterfly ray for a common stingray, a rare find that fellow enthusiast Brian Sng identified from a photograph. To prevent such errors, the developers included a pop-up feature that alerts users when a scanned fish is a protected, rare, or harmful species, advising them to release endangered catches. The app displays the top five predictions ranked by similarity so users can finalise the correct species identity. For field convenience, it runs a lightweight offline model when mobile connections are weak, alongside a more accurate online version. Guan Chen Di, a 22-year-old computer science student at Nanyang Technological University who worked on the app, noted that fitting all necessary information onto a tiny phone screen presented a major design hurdle.

Citizen Science Helps Supplement Traditional Marine Surveys

Fish biologist Jeffrey Kwik stated that citizen science initiatives like SGFish help supplement traditional fish surveys involving diving or snorkelling, which depend heavily on good water visibility. Spotting fish populations visually in Singapore is particularly challenging due to murky waters and well-camouflaged species hiding in seagrass and corals, according to the Singapore Institute of Technology associate professor. Andriel Cheong, co-founder of the Ichthyological Society of Singapore, hopes the database will eventually inform environmental impact assessments for coastal developments. The developers plan to release the app on the Google Play Store in future updates and incorporate gamification elements from the original Telegram bot, such as a high-score board recognizing users who record the highest species diversity.