In the ex situ conservation and population management of giant pandas (Ailuropoda melanoleuca), timely and rapid individual identification and behavior monitoring play a crucial role in health supervision. The health status of captive giant pandas is usually observed by specialized breeders, which have high labor costs, low efficiency, and lack of timeliness. Therefore, efficient and cheap image-based identity recognition and behavior analysis has become a new development trend. Existing studies have proposed to realize it by detecting and analyzing facial images of giant pandas. However, this type of method still has the problem of insufficient detection accuracy, which makes it difficult to improve the recognition performance. We propose a dual model fusion method based on YOLOv3 and Mask R-CNN to achieve the segmentation and accurate detection of giant panda head images. It includes three parts:YOLOv3 completes head detection, Mask R-CNN completes panda contour segmentation, and then fuses the outputs of the two models. Experimental results show that the accuracy of head detection, giant panda and head contour segmentation are 82. 6%, 95. 2%, and 87. 1%, respectively. Our method has high detection and segmentation accuracy for the giant panda head images. It provides help for individual identification and gender classification of giant pandas and provides technical reference for behavior analysis.
Chen J, Wen Q, Zhuo C L, Mete M. 2012a. A novel approach towards head detection of giant pandas in the free-range environment. Chongqing:International Congress on Image and Signal Process-ing, 814-818.
Chen J, Wen Q, Zhuo C L, Mete M. 2012b. Pose recognition of giant pandas based on gradient shapes. Leshan:International Confer-ence on Computational Problem-Solving, 358-362.
Chen J, Wen Q, Qu W M, Mete M. 2012c. Panda facial region detec-tion based on topology modelling. Chongqing:International Con-gress on Image and Signal Processing, 911-915.
Chen P, Swarup P, Matkowski W M, Kong A W K, Han S, Zhang Z, Rong H. 2020. A study on giant panda recognition based on im-ages of a large proportion of captive pandas. Ecology and Evolu-tion, 10 (7):3561-3573.
Ding Q Q, Yu J H, Xie C Y, Xu C Z, Sun Q, Bi J Z, Chu Q P, Zhou Y M, Wu F. 2019. A preliminary study on the maternal behavior of female giant pandas in Shanghai Wild Animal Park. Acta Therio-logica Sinica, 39 (3):302-310. (in Chinese)
He K M, Gkioxari G, Dollar P, Girshick R. 2020. Mask R-CNN. Transactions on Pattern Analysis & Machine Intelligence, 42 (2):386-397.
He K M, Zhang X Y, Ren S Q, Sun J. 2016. Deep residual learning for image recognition. Las Vegas:Conference on Computer Vi-sion and Pattern Recognition (CVPR), 770-778.
He Y G, Song S X, Ma K, Feng L, Li D S, Liu D Z. 2015. Translocat-ing two lactating giant pandas and their cubs after behavioral training did not impair maternal behavior:a case study. Acta The-riologica Sinica, 35 (3):342-347. (in Chinese)
Hou J, He Y X, Yang H B, Connor T, Gao J, Wang Y J, Zeng Y C, Zhang J D, Huang J Y, Zheng B C, Zhou S Q. 2020. Identifica-tion of animal individuals using deep learning:A case study of gi-ant panda. Biological Conservation, 242. DOI:10. 1016/j. bio-con. 2020. 108414.
Kühl S H, Burghardt T. 2013. Animal biometrics:quantifying and de-tecting phenotypic appearance. Trends in Ecology & Evolution(Amsterdam), 28 (7):432-441.
Lin T Y, Dollár P, Girshick R, He K, Hariharan B, Belongie S. 2017. Feature pyramid networks for object detection. Honolulu:Confer-ence on Computer Vision and Pattern Recognition (CVPR), 936-944.
Redmon J, Divvala S, Girshick R, Farhadi A. 2016. You only look once:unified, real-time object detection. Las Vegas:Conference on Computer Vision and Pattern Recognition (CVPR), 779-788.
Redmon J, Farhadi A. 2018. YOLOv3:An Incremental Improvement. https://arXiv.org/abs/1804.02767.
Ren S, He K, Girshick R, Sun J. 2017. Faster R-CNN:towards realtime object detection with region proposal networks. Transac-tions on Pattern Analysis and Machine Intelligence, 39 (6):1137-1149.
Swarup P, Chen P, Hou R, Que P J, Liu P, Kong A W K. 2021. Giant panda behaviour recognition using images. Global Ecology and Conservation, 26:e01510.
Wang H N, Su H, Chen P, Hou R, Zhang Z H, Xie W Y. 2019. Learn-ing deep features for giant panda gender classification using face images. Seoul:International Conference on Computer Vision Workshop (ICCVW), 27-28.
Yang K X, Peng W P, Zhang H, Zhuo G F, Wu K J, Yin H, Yuan B, Chen M, Xie Y. 2019. Research on the breeding of captive red panda (Ailurus fulgens). Chinese Journal of Wildlife, 40 (1):177-181. (in Chinese)
Zhang B W, Li M, Zhang Z J, Goossens B, Zhu L F, Zhang S N, Hu J C, Bruford M W, Wei F W. 2007. Genetic viability and popula-tion history of the giant panda, putting an end to the "evolutionary dead end"? Molecular Biology and Evolution, 24 (8):1801-1810.
Zhou Z. 2019. Research on method of reading time of pointer watch based on convolutional neural network. Master thesis. Changsha:Hunan University. (in Chinese)
丁乔棋, 俞锦华, 谢春雨, 徐春忠, 孙强, 毕靖征, 褚青坡, 周应敏, 吴锋. 2019. 上海野生动物园大熊猫育幼行为初步研究. 兽类学报, 39 (3):302-310.
杨奎兴, 彭文培, 张皓, 卓桂富, 吴孔菊, 尹洪, 袁博, 陈敏, 谢意. 2019. 圈养小熊猫繁殖初探. 野生动物学报, 40 (1):177-181.
何永果, 宋仕贤, 马凯, 冯莉, 李德生, 刘定震. 2015. 哺乳期圈养雌性大熊猫圈舍转移及其育幼行为变化. 兽类学报, 35 (3):342-347.
周珍. 2019. 基于卷积神经网络的指针手表读时方法研究. 长沙:湖南大学硕士学位论文.
黄佳琪, 王雪琴, 曾陈颖, 肖江. 2014. 基于灰度空间大熊猫图像分割方法研究. 科技资讯, 11 (7):11-14.
青公网安备 63010402000199号 青ICP备05000010号-2
