ALGORITHM FOR THE DETECTION OF URBAN TREES FROM 360 IMAGES
Keywords:
Tree Detection; YOLO; Equirectangular Image; Image 360; Computer Vision.Abstract
Trees are indispensable to human life, they absorb carbon dioxide and release oxygen, help moderate temperature, protect ecosystems, and reduce erosion. The manual identification of trees on public roads requires expense and time for recording and managing the data collected, since urban regions can be very large. We developed in this paper a method for the trees recognition and identification in urban areas from a 360 video. A YOLO neural network was trained to detect the trees from frames of the equirectangular video (360 images). We used Computer Vision techniques with the OpenCV library to develop algorithms to segment the regions that fit the detected trees in the rectilinear field of view (gnomonic projection), in order to verify if the trees are on the sidewalks. The results obtained showed around 80% success in detecting trees using YOLO, and an accuracy of 71% in the algorithm that checks if the trees are on the sidewalk.
Downloads
References
ALVES, G.; Detecção de Objetos com YOLO - Uma abordagem moderna. 2020. Disponível em: https://iaexpert.academy/2020/10/13/deteccao-de-objetos-com-yolo-uma-abordagem-moderna. Acesso em: 17 dez. 2022.
BARBOSA, R. L.; GALLIS, R. B. A.; HIRAGA, A. K.; SILVA, F. A. Quantificação e Georreferenciamento Semiautomático de Árvores Urbanas. Revista da Sociedade Brasileira de Arborização Urbana - REVSBAU, Curitiba, v.13, n.4, p.41-53, 2018. https://doi.org/10.5380/revsbau.v13i4.65046
BOURAYA, S.; BELANGOUR, A. Deep Learning based Neck Models for Object Detection: A Review and a Benchmarking Study. International Journal of Advanced Computer Science and Applications. 2021. Acesso em: 14 dez. 2022. https://doi.org/10.14569/IJACSA.2021.0121119
GLEDHILL, D.; TIAN, G. Y.; TAYLOR, D.; CLARKE, D. Panoramic imaging—a review. Computer & Graphics, v. 27, n. 3, p. 435-445, 2003. https://doi.org/10.1016/S0097-8493(03)00038-4
IBFLORESTAS; Quais são as partes da árvore e as suas funções?. s.d. Disponível em: https://www.ibflorestas.org.br/conteudo/quais-sao-as-partes-da-arvore?utm_source=google-ads&utm_medium=cpc&utm_campaign=nativas-c&keyword=%22a%20importancia%20da%20arvore%27&creative=429664467327&gclid=Cj0KCQjwnoqLBhD4ARIsAL5JedLWmXY4yn5p5yCqHCmlb8VdrncdeSzJeWCaaDI-sgT8VQVOighB6_YaAsDoEALw_wcB. Acesso em: 03 dez. 2022.
ITAKURA, K.; HOSOI, F. Automatic Tree Detection from Three-Dimensional Images Reconstructed from 360° Spherical Camera Using YOLO v2. Remote Sensing, 12(6), 2020. https://doi.org/10.3390/rs12060988
KATEB, F. A.; MONOWAR, M. M.; HAMID, A.; OHI, A. Q.; MRIDHA, M. F. FruitDet: Attentive Feature Aggregation for Real-Time Fruit Detection in Orchards. Agronomy. 11(12), 2021. https://doi.org/10.3390/agronomy11122440
MUTHA, N. How to map Equirectangular projection to Rectilinear projection. 2017. Disponível em: http://blog.nitishmutha.com/equirectangular/360degree/2017/06/12/How-to-project-Equirectangular-image-to-rectilinear-view.html. Acesso em: 11 dez. 2022.
OPEN IMAGES DATASET. Disponível em: https://storage.googleapis.com/openimages/web/index.html. Acesso em: 19 dez. 2022.
OTSU, N. A Threshold Selection Method from Gray-Level Histograms. IEEE Transactions on Systems, Man, And Cybernetics, v. 9, n. 1, p. 62-66, 1979.. https://doi.org/10.1109/TSMC.1979.4310076
PIEMONTEZ; YOLO para Detecção de Objetos – Visão Geral. 2022. Disponível em: https://visaocomputacional.com.br/yolo-para-deteccao-de-objetos-visao-geral/. Acesso em: 13 dez. 2022.
RAJPUT. Working of YOLOv4 Algorithm. 2021. Disponível em: https://medium.com/@shraddhapattanshetti161998/working-of-yolov4-algorithm-76a5e75f188b#:~:text=YOLOv4%20Specifically%20uses%20CSPDarknet53%20as. Acesso em: 09 dez. 2022.
SHINDE, S.; KOTHARI, A.; GUPTA, V. YOLO based Human Action Recognition and Localization. Procedia Computer Science, 133, p. 831–838. 2018. https://doi.org/10.1016/j.procs.2018.07.112
SOLAWETZ; YOLOv4 - An explanation of how it works. 2020. Disponível em: https://blog.roboflow.com/a-thorough-breakdown-of-yolov4/#yolov4-backbone-network-feature-formation. Acesso em: 10 dez. 2022.
SUZUKI, S.; BE, K. Topological structural analysis of digitized binary images by border following. Computer Vision, Graphics, and Image Processing, 30(1), 32–46, 1985. https://doi.org/10.1016/0734-189X(85)90016-7
TREIBER, M. An Introduction to Object Recognition: Selected Algorithms for a Wide Variety of Applications. Springer-Verlag London Limited, 2010.
WEISSTEIN. Gnomonic Projection. s.d. Disponível em: https://mathworld.wolfram.com/GnomonicProjection.html. Acesso em: 13 dez. 2022.
XIE, Q.; LI, D.; YU, Z.; ZHOU, J.; WANG, J. Detecting Trees in Street Images via Deep Learning with Attention Module. IEEE Transactions on Instrumentation and Measurement, v. 69, n. 8, p. 5395-5406, 2019. https://doi.org/10.1109/TIM.2019.2958580