Japan Turns to AI to Prevent Tree-Related Accidents in Cities

Jakarta — Engineers in Japan are advancing an artificial intelligence (AI) system designed to identify trees at risk of collapsing or shedding dangerous branches, as authorities grapple with aging urban greenery and rising public safety concerns.

According to Kyodo News, several local governments have begun exploring the adoption of the technology, citing a shortage of trained arborists responsible for inspecting tree health in parks and along roadways.

The current system is capable of assessing the condition of zelkova and cherry blossom (sakura) trees, with developers planning to expand its capabilities to include ginkgo and other species in the near future.

Data from Japan’s Ministry of Land, Infrastructure, Transport and Tourism recorded 1,732 tree-related accidents in parks and roadside areas between April 2021 and November 2024. Of these incidents, 110 resulted in injuries or fatalities.

One notable case occurred in September 2024, when a man died after being struck by a falling ginkgo branch on a pedestrian path in Hino, western Tokyo.

Tokyo-based IT firm Optim Corp., in collaboration with Kyuden Droneservice Co. in Fukuoka, has developed a drone-powered AI system capable of detecting dead or deteriorating trees from above.

Meanwhile, Sumitomo Mitsui Construction Co. has introduced an AI-based risk assessment tool that analyzes photographs of tree trunks and decaying sections captured via smartphones or tablets.

The system, known as Tree AI, compares submitted images against tens of thousands of stored photographs of tree trunks and branches, alongside diagnostic data from certified arborists. It then categorizes the likelihood of collapse into four risk levels.

The analysis can be completed within minutes, with results automatically stored in a centralized database that includes geolocation data for each tree.

“Collapse risks increase when trees develop internal cavities or fungal growth within the trunk, conditions that are difficult for untrained individuals to detect,” said tree specialist Hidemi Kataoka, who is involved in the system’s development.

Several municipalities—including Kizugawa in Kyoto Prefecture, Miyakonojo in Miyazaki Prefecture, and the Tokyo Metropolitan Government—have begun pilot testing the technology.

Kizugawa, which manages approximately 20,000 trees, has acknowledged challenges in conducting routine inspections due to limited personnel.

“With AI, inspections can become more efficient and consistent,” a Kizugawa official said.

Around 20 municipalities across Japan are currently considering implementing the system. Sumitomo Mitsui Construction aims to launch the AI service commercially by fiscal year 2027.

“Many trees were planted during Japan’s period of rapid economic growth to absorb vehicle emissions, and they are now reaching the end of their lifespan,” said Sho Tago, head of the company’s Green Infrastructure Division, expressing hope that the technology will support more efficient tree management amid labor shortages.

Indonesianpost.com | Kabarin