アフリカゴールデンキャットの個体数こたいすう調査ちょうさ始はじまる　AIと画像がぞう認識にんしきも活用かつよう

When zoologist Mwezi Mugerwa first saw the African golden cat, he didn’t know what kind of animal it was.

When Mr. Mugerwa checked the footage from cameras set up in the forests of Uganda in the database, he discovered a strange animal, about twice the size of a domestic cat, appearing in the blurry black-and-white images.

This mysterious yellow cat is called Mbaka in the local language and is often caught in traps set in the forest.

This species may become extinct before we even understand it, said Mr. Mugerwa.

For the past 16 years, Mr. Mugerwa has dedicated his career to this large feline, the least known, understood, and studied in Africa.

It is believed that this creature inhabits the dense forests of Central and West Africa, but it is extremely difficult to spot.

Even the International Union for Conservation of Nature IUCN did not record an estimated population in its last assessment over 10 years ago.

Mr. Mugelwa himself has only witnessed it three times, despite many years of field research.

It is really very difficult to find this species in the wild.

Many of their habitats are located in extremely remote and isolated areas, said Mr. Mugerwa.

This year, Mr. Mugerwa was recognized for his efforts regarding the African golden cat and received the Emerging Conservationist Award at the Indianapolis Prize.

Recognizing that accurately counting the population is the first step in conservation, Mr. Mugerwa began the first population surveys in various habitats.

The results are scheduled to be announced next year.

Research using camera traps has become common in the study of wild cats, and researchers identify individuals by their distinctive spots and stripes.

However, since each photo must be checked individually, there are also limitations to installing camera traps.

Therefore, in 2019, he proposed the idea of building a network.

Although it took a lot of time to persuade my colleagues, there are now 46 experts from 19 countries participating in the network.

This network is called the African Golden Cat Conservation Alliance AGCCA, and experts have started camera trap surveys throughout the entire range believed to be the habitat of the African golden cat.

The image is provided by the National Geographic Society.

However, checking thousands of photos from camera traps set up in 30 regions across 19 countries is by no means an easy task.

It was really very difficult, recalls Mr. Mugerwa.

Around the same time, his partner, the U.S.-based nonprofit organization Panthera, developed an AI algorithm that can identify individuals based on their distinctive patterns and quickly classify images.

This method is similar to using a tiger’s stripes like fingerprints.

This is really important.

Thanks to that, we are now able to talk about the number and density of individuals, Mr. Mugerwa said.

He believes that it is almost impossible to identify individuals without AI.

According to provisional data, the population density of the African golden cat is extremely low even within protected areas.

In Uganda and Gabon, only 16 individuals have been recorded per 100 square kilometers.

The chain of investigations also reveals the actual impact of poaching.

According to Mr. Mugelwa, the density and distribution range are up to 50% higher in areas where hunting is prohibited.

According to research, the African golden cat is active both day and night, but most individuals are strictly nocturnal.

Its probably to avoid human activity during the day.