DETECTION OF DEER IN IMAGES TO ACCOUNT POPULATIONS BY COMPUTER VISION METHODS
Abstract
Purpose. Recognition of deer individuals by their graphic images based on convolutional neural network models.
Research methods and materials. The material for the research was marked images of deer in different habitats, published in the public domain. Research methods: theory of design and development of artificial intelligence systems; image augmentation for computer vision tasks; hyperparameter tuning algorithms for training neural network models.
Results. Conservation of the deer population is one of the important goals in ecology and agriculture. The control of the number of individuals affects the conservation of rare species of deer, and supports the production volumes on farms. For example, such a control method may be the detection of deer individuals in photographs. To accurately determine and count the number of deer, you can use a tool such as a neural network.
We used deep learning methods for convolutional neural networks, as well as the concept of “transfer learning”. On the basis of the Faster R-CNN Resnet50 network, a neural network was trained, which allows, with an accuracy of 0.91, to determine individual individuals in the images using the F1-score metric with a threshold value of 0.6 (coincidence of the area of the predicted markup and the actual one) with an accuracy of 0.91.
Conclusion. To solve the problem of deer detection, a data set has been prepared, including more than 30 thousand images, with markings of individual individuals. The marking of each individual included the coordinates of the bounding rectangle in the image. Based on this information, a neural network model was developed and trained using software tools to solve the problem of detecting deer objects in images. The experiments carried out showed that the accuracy of detection with augmentation in accordance with the F1-score indicator with a threshold value of 0.6 on the training sample was 0.96, and on the test sample was 0.91.
Downloads
References
Список литературы
Понимание концепций IU и IoU (Intersection over Union) в глубоком обучении и реализации программ на Python / Русские Блоги. URL: https://russianblogs.com/article/4211133576/ (дата обращения: 22.03.2022).
Современная история таймырской популяции дикого северного оленя: динамика, управление, угрозы и пути сохранения / Серия Экологические исследования. URL: http://journals.krc.karelia.ru/index.php/ecology/article/view/1045 (дата обращения: 22.03.2022).
Терещенко С.Н., Перов А.А., Осипов А.Л. Определение заболеваний маниока методами компьютерного зрения // Siberian Journal of Life Sciences and Agriculture. 2021. Т. 13. № 1. С. 144-155. https://doi.org/10.12731/2658-6649-2021-13-1-144-155
Терещенко С.Н., Перов А.А., Осипов А.Л. Классификация внекорневых заболеваний яблоневых культур методами компьютерного зрения // Siberian Journal of Life Sciences and Agriculture. 2021. Т. 13. № 3. С. 103-118. https://doi.org/10.12731/2658-6649-2021-13-3-103-118
Численность охотничьих ресурсов / Федеральное государственное бюджетное учреждение "Федеральный центр развития охотничьего хозяйства". URL: http://www.ohotcontrol.ru/resource/number/ (дата обращения: 22.03.2022).
Region of interest pooling explained / Deepsense.ai. URL: https://deepsense.ai/region-of-interest-pooling-explained/ (дата обращения: 22.03.2022).
Kilgo J.C., Ray H.S., Vukovich M., Goode M.J., Ruth C. Predation by coyotes on white-tailed deer neonates in South Carolina // Journal of Wildlife Management, 2012, vol. 76(7), pp. 1420-1430. https://doi.org/10.1002/jwmg.393
Loe L.E., Mysterud A., Veiberg V., Langvatn R. Negative density-dependent emigration of males in an increasing red deer population // Proc Biol Sci., 2009, vol. 276(1667), pp. 2581-2587. https://doi.org/10.1098/rspb.2009.0224
Lyubing Zhang, Eric I. Ameca y Juárez, Zhigang Jiang. Viability analysis of the wild sika deer (Cervus nippon) population in China: Threats of habitat loss and effectiveness of management interventions // Journal for Nature Conservation, 2018, vol. 43, pp. 117-125. https://doi.org/10.1016/j.jnc.2018.02.014
Mariana B. Nagy-Reis, Mark A. Lewis, William F. Jensen, Mark S. Boyce. Conservation Reserve Program is a key element for managing white-tailed deer populations at multiple spatial scales // Journal of Environmental Management, 2019, vol. 248. https://doi.org/10.1016/j.jenvman.2019.109299
Marrett D. Grund, Alan Woolf. Development and evaluation of an accounting model for estimating deer population sizes // Ecological Modelling, 2004, vol. 180, no. 2-3, pp. 345-357. https://doi.org/10.1016/j.ecolmodel.2004.04.025
Region Proposal Network (RPN) — Backbone of Faster R-CNN. URL: https://medium.com/egen/region-proposal-network-rpn-backbone-of-faster-r-cnn-4a744a38d7f9 (дата обращения: 22.03.2022).
Michelle H.G. Wong Yanni Mo, Bosco Pui Lok Chan. Past, present and future of the globally endangered Eld’s deer (Rucervus eldii) on Hainan Island, China // Global Ecology and Conservation, 2021, vol. 26. https://doi.org/10.1016/j.gecco.2021.e01505
Challenges / Reindeer Herding. URL: https://reindeerherding.org/challenges (дата обращения: 22.03.2022).
Herders / Reindeer Herding. URL: https://reindeerherding.org/herders (дата обращения: 22.03.2022).
Tereshchenko S.N., Moiseeva E.D. Computer vision for determining the fraction of proppant for hydraulic fracturing // Proceedings - 2021 International Conference on Industrial Engineering, Applications and Manufacturing, ICIEAM, 17-21 May 2021, pp. 742–746. https://doi.org/10.1109/ICIEAM51226.2021.9446381
Tereshchenko S.N., Osipov A.L., Moiseeva E.D. Determining the Number of Wheatears on the Images of Wheat Fields by the Computer Vision Method // Optoelectronics, Instrumentation and Data Processing, 2022, vol. 58(3), pp. 266-272. https://doi.org/10.3103/S8756699022030074
Tereshchenko S.N., Perov A.A., Osipov A.L. Features of Applying Pretrained Convolutional Neural Networks to Graphic Image Steganalysis // Optoelectronics, Instrumentation and Data Processing, 2021, vol. 57(4), pp. 419-425. https://doi.org/10.3103/S8756699021040117
Faster RCNN Object detection. URL: https://towardsdatascience.com/faster-rcnn-object-detection-f865e5ed7fc4 (дата обращения: 22.03.2022).
Zini V., Wäber K., Hornigold K. et al. Human and environmental associates of local species-specific abundance in a multi-species deer assemblage // Eur J Wildl Res, 2021, vol. 67, 99. https://doi.org/10.1007/s10344-021-01539-6
References
Understanding the concepts of IU and IoU (Intersection over Union) in deep learning and program implementation in Python / Russian Blogs. URL: https://russianblogs.com/article/4211133576/ (accessed 22.03.2022).
Modern history of the Taimyr wild reindeer population: dynamics, management, threats and ways of conservation / Ecological Research Series. URL: http://journals.krc.karelia.ru/index.php/ecology/article/view/1045 (accessed 22.03.2022).
Tereshchenko S.N., Perov A.A., Osipov A.L. Determination of cassava diseases by computer vision methods. Siberian Journal of Life Sciences and Agriculture, 2021, vol. 13, no. 1, pp. 144-155. https://doi.org/10.12731/2658-6649-2021-13-1-144-155
Tereshchenko S.N., Perov A.A., Osipov A.L. Classification of foliar diseases of apple crops by computer vision methods. Siberian Journal of Life Sciences and Agriculture, 2021, vol. 13, no. 3, pp. 103-118. https://doi.org/10.12731/2658-6649-2021-13-3-103-118
The number of hunting resources / Federal State Budgetary Institution “Federal Center for the Development of Hunting Economy”. URL: http://www.ohotcontrol.ru/resource/number/ (accessed 22.03.2022).
Region of interest pooling explained / Deepsense.ai. URL: https://deepsense.ai/region-of-interest-pooling-explained/ (accessed 22.03.2022).
Kilgo J.C., Ray H.S., Vukovich M., Goode M.J., Ruth C. Predation by coyotes on white-tailed deer neonates in South Carolina. Journal of Wildlife Management, 2012, vol. 76(7), pp. 1420-1430. https://doi.org/10.1002/jwmg.393
Loe L.E., Mysterud A., Veiberg V., Langvatn R. Negative density-dependent emigration of males in an increasing red deer population. Proc Biol Sci., 2009, vol. 276(1667), pp. 2581-2587. https://doi.org/10.1098/rspb.2009.0224
Lyubing Zhang, Eric I. Ameca y Juárez, Zhigang Jiang. Viability analysis of the wild sika deer (Cervus nippon) population in China: Threats of habitat loss and effectiveness of management interventions. Journal for Nature Conservation, 2018, vol. 43, pp. 117-125. https://doi.org/10.1016/j.jnc.2018.02.014
Mariana B. Nagy-Reis, Mark A. Lewis, William F. Jensen, Mark S. Boyce. Conservation Reserve Program is a key element for managing white-tailed deer populations at multiple spatial scales. Journal of Environmental Management, 2019, vol. 248. https://doi.org/10.1016/j.jenvman.2019.109299
Marrett D. Grund, Alan Woolf. Development and evaluation of an accounting model for estimating deer population sizes. Ecological Modelling, 2004, vol. 180, no. 2-3, pp. 345-357. https://doi.org/10.1016/j.ecolmodel.2004.04.025
Region Proposal Network (RPN) — Backbone of Faster R-CNN. URL: https://medium.com/egen/region-proposal-network-rpn-backbone-of-faster-r-cnn-4a744a38d7f9 (accessed 22.03.2022).
Michelle H.G. Wong Yanni Mo, Bosco Pui Lok Chan. Past, present and future of the globally endangered Eld’s deer (Rucervus eldii) on Hainan Island, China. Global Ecology and Conservation, 2021, vol. 26. https://doi.org/10.1016/j.gecco.2021.e01505
Challenges / Reindeer Herding. URL: https://reindeerherding.org/challenges (accessed 22.03.2022).
Herders / Reindeer Herding. URL: https://reindeerherding.org/herders (accessed 22.03.2022).
Tereshchenko S.N., Moiseeva E.D. Computer vision for determining the fraction of proppant for hydraulic fracturing. Proceedings - 2021 International Conference on Industrial Engineering, Applications and Manufacturing, ICIEAM, 17-21 May 2021, pp. 742–746. https://doi.org/10.1109/ICIEAM51226.2021.9446381
Tereshchenko S.N., Osipov A.L., Moiseeva E.D. Determining the Number of Wheatears on the Images of Wheat Fields by the Computer Vision Method. Optoelectronics, Instrumentation and Data Processing, 2022, vol. 58(3), pp. 266-272. https://doi.org/10.3103/S8756699022030074
Tereshchenko S.N., Perov A.A., Osipov A.L. Features of Applying Pretrained Convolutional Neural Networks to Graphic Image Steganalysis. Optoelectronics, Instrumentation and Data Processing, 2021, vol. 57(4), pp. 419-425. https://doi.org/10.3103/S8756699021040117
Faster RCNN Object detection. URL: https://towardsdatascience.com/faster-rcnn-object-detection-f865e5ed7fc4 (accessed 22.03.2022).
Zini V., Wäber K., Hornigold K. et al. Human and environmental associates of local species-specific abundance in a multi-species deer assemblage. Eur J Wildl Res, 2021, vol. 67, 99. https://doi.org/10.1007/s10344-021-01539-6
Copyright (c) 2024 Sergei N. Tereshchenko, Alexander L. Osipov, Ekaterina D. Moiseeva
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
