Detecting tree and wire entanglements with deep learning

• Published in: Trees (Berlin, West) West), 2023-02, Vol.37 (1), p.147-159
• Main Authors: Oliveira, Artur André, Buckeridge, Marcos S., Hirata, Roberto
• Format: Article
• Language: English
• Subjects: Agriculture; Air pollution; Biomedical and Life Sciences; Communication; Deep learning; Electric contacts; Forestry; Life Sciences; Plant Anatomy/Development; Plant Pathology; Plant Physiology; Plant Sciences; Pollution control; Qualitative analysis; Review; Trees; Urban heat islands; Urban Trees; Wiring
• ISSN: 0931-1890; 1432-2285
• DOI: 10.1007/s00468-022-02305-0

Power and communication line corridors are usually mixed with urban trees, and this mixing can be the source of multiple issues like fires and communication failures. Nevertheless, urban trees are a valuable resource to the city as they dissipate heat island effects, reduce air pollution and increase general health perception. This work proposes a deep learning approach to detect trees entangled to power and communication lines using street-level imagery and perform quick quantitative and qualitative analyses based on the Grad-CAM++ method. Testing the method was performed using 1001 images from urban trees from the cities of São Paulo and Porto Alegre (both in Brazil). We found an overall accuracy of 74.6% (73.6% for São Paulo and 75.6% for Porto Alegre), suggesting that the methodology could be suitable in the future for city management to avoid risks of accidents due to contact between trees and electrical wiring. This text describes the method, a new data set of urban images, the experimental setup design and tests, and some possible future improvements.