基于Faster R-CNN金丝猴优化检测方法

采用深度学习方法,利用网络爬取、实地拍摄两种方式获取数据,并构建基于Faster R-CNN(region-based convolutional neural networks)的金丝猴优化检测模型。通过比较不同的模型,明确模型的最优构建方案;通过对比基于不同训练集构建的模型的检测精度,探究建模数据最优补充方案。研究结果表明:相比Vgg16和Res50网络,基于Res101网络在迭代70000次时可以构建最优的模型;在实测数据有限时,可以采用网络图片作为替代数据源进行金丝猴面部检测,并且可以将其作为辅助数据源优化金丝猴身体检测效果;相对于经典的卷积神经网络,该方法不仅检测效果更好,而且运行时间更少;该模型可以在复杂生态背景图片中有效地进行金丝猴定位与识别。所提方法对金丝猴野外发现与跟踪有很强的现实意义。

Abstract

In this paper, deep learning method is adopted to obtain data by network crawling and field shooting, and an optimized detection model of snub-nosed monkeys based on Faster region-based convolutional neural networks (R-CNN) is constructed. An optimal modeling scheme is determined by comparing different models, and the detection precision of different models constructed based on different training datasets is compared to explore the optimal supplementary scheme for modeling data. The results indicate that, compared with Vgg16 and Res50 networks, the optimal model can be constructed based on Res101 network with 70000 iterations. The network images could be used as an alternative data source for face detection of snub-nosed monkeys, and the network images could also be used as an auxiliary data source to optimize the body detection effects of snub-nosed monkeys when the measured data is limited. Compared with classical convolutional neural networks, this method has better detection effect and less running time. Among the images containing the complicate eco-environmental background information, this model could effectively locate and identify snub-nosed monkeys. The proposed algorithm has a strong practical significance for the discovery and tracking of snub-nosed monkeys in the wild.

参考文献

[1] 李明, 向左甫. 中国旗舰保护物种: 金丝猴[J]. 世界环境, 2016( S1): 12- 15.

LiM, Xiang ZF. Flagship species under China's protection: golden monkeys[J]. World Environment, 2016( S1): 12- 15.

[2] . 金丝猴生态的初步研究[J]. 动物学研究, 1982, 3(2): 105-110.

. Preliminary studies on the ecology of the golden-haired monkey[J]. Zoological Research, 1982, 3(2): 105-110.

[3] , et al. Genetic diversity of the Sichuan snub-nosed monkey (Rhinopithecus roxellana) in Shennongjia National Park, China using RAD-seq analyses[J]. Genetica, 2019, 147(3/4): 327-335.

[4] , et al. Epigenetic and transcriptional signatures of ex situ conserved golden snub-nosed monkeys (Rhinopithecus roxellana)[J]. Biological Conservation, 2019, 237: 175-184.

[5] 穆俊明. 神农架保护区金丝猴生境退化机制及恢复技术[D]. 武汉: 华中农业大学, 2015: 34- 35.

Mu JM. The habitat degradation and recovery technology of the golden snub-nosed monkey in Shennongjia nature reserve[D]. Wuhan: Huazhong Agricultural University, 2015: 34- 35.

[6] , 等. 湖北川金丝猴现状及保护研究[J]. 野生动物学报, 2019, 40(3): 602-609.

, et al. Protection and status of Sichuan snub-nosed monkey in Hubei Province[J]. Chinese Journal of Wildlife, 2019, 40(3): 602-609.

[7] , et al. Congruence between arthropod and plant diversity in a biodiversity hotspot largely driven by underlying abiotic factors[J]. Ecological Applications, 2019, 29(4): e01883.

[8] . The effects of anthropogenic landscape change on the abundance and habitat use of terrestrial large mammals of Nech Sar National Park[J]. Environmental Systems Research, 2019, 88(1): 1-16.

[9] , 等. 图像理解中的卷积神经网络[J]. 自动化学报, 2016, 42(9): 1300-1312.

, et al. Convolutional neural networks in image understanding[J]. Acta Automatica Sinica, 2016, 42(9): 1300-1312.

[10] 方楠. 基于CNN的金丝猴面部识别研究与实现[D]. 西安: 西安电子科技大学, 2017: 36- 37.

FangN. Research and implement of gold monkey recognition based on convolutional neural network[D]. Xi'an:Xidian University, 2017: 36- 37.

[11] 张飞飞. 复杂背景下的多姿态猴脸检测与识别方法研究[D]. 西安: 西安理工大学, 2018: 11- 12.

Zhang FF. Research on methods of detection and recognition of multi-pose monkey face in complex background[D]. Xi'an: Xi'an University ofTechnology, 2018: 11- 12.

[12] 范莹莹. 基于深度学习的金丝猴面部识别软件设计与实现[D]. 西安: 西安电子科技大学, 2018: 35- 36.

Fan YY. Design and implementation of golden monkey face recognition software based on deep learning[D]. Xi'an:Xidian University, 2018: 35- 36.

[13] 王革伟. 川金丝猴面部识别方法研究[D]. 西安: 西北大学, 2018: 27- 28.

Wang GW. The research on face recognition of Sichuan golden monkey[D]. Xi'an:Northwest University, 2018: 27- 28.

[14] , et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.

[15] LanG, YoshuaB, AaronC. Deep learning[M]. Beijing: Posts & Telecom Press, 2017: 412- 413.

[16] , 等. 蝴蝶种类自动识别研究[J]. 计算机研究与发展, 2018, 55(8): 1609-1618.

, et al. The automatic identification of butterfly species[J]. Journal of Computer Research and Development, 2018, 55(8): 1609-1618.

[17] , 等. 基于Faster R-CNN的榆紫叶甲虫识别方法研究[J]. 计算机工程与应用, 2018, 54(23): 89-93, 108.

, et al. Identification method of ambrostoma quadriimpressum motschlsky based on Faster R-CNN[J]. Computer Engineering and Applications, 2018, 54(23): 89-93, 108.

[18] GirshickR, DonahueJ, DarrellT, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]∥2014 IEEE Conference on Computer Vision and Pattern Recognition, June 23-28, 2014, Columbus, OH, USA. IEEE, 2014: 580- 587.

[19] GirshickR. Fast R-CNN[C]∥2015 IEEE International Conference on Computer Vision (ICCV), December 7-13, 2015, Santiago, Chile. IEEE, 2015: 1440- 1448.

[20] , et al. Deep learning model design of video target tracking based on TensorFlow platform[J]. Laser & Optoelectronics Progress, 2017, 54(9): 091501.

[21] . Hoi S C H. Face detection using deep learning: an improved faster RCNN approach[J]. Neurocomputing, 2018, 299: 42-50.

[22] , et al. Mean Normalized Retrieval Order (MNRO): a new content-based image retrieval performance measure[J]. Multimedia Tools and Applications, 2014, 70(3): 1767-1798.

[23] , 等. 基于改进的Faster R-CNN目标检测算法[J]. 激光与光电子学进展, 2019, 57(10): 101009.

, et al. Based on improved Faster R-CNN target detection algorithm[J]. Laser & Optoelectronics Progress, 2019, 57(10): 101009.

[24] . 基于深度学习的目标检测与可行域分割研究[J]. 激光与光电子学进展, 2019, 57(12): 121013.

. Research on target detection and feasible region segmentation based on deep learning[J]. Laser & Optoelectronics Progress, 2019, 57(12): 121013.

[25] , 等. 基于Faster R-CNN深度网络的油菜田间杂草识别方法[J]. 激光与光电子学进展, 2019, 57(2): 021508.

, et al. Identification method of weeds in rapeseed field based on Faster R-CNN deep network[J]. Laser & Optoelectronics Progress, 2019, 57(2): 021508.

孙蕊, 张旭, 郭颖, 于新文, 陈艳, 侯亚男. 基于Faster R-CNN金丝猴优化检测方法[J]. 激光与光电子学进展, 2020, 57(12): 121022. 孙蕊, 张旭, 郭颖, 于新文, 陈艳, 侯亚男. Optimized Detection Method for Snub-Nosed Monkeys Based on Faster R-CNN[J]. Laser & Optoelectronics Progress, 2020, 57(12): 121022.

基于Faster R-CNN金丝猴优化检测方法

关于本站 Cookie 的使用提示

全站搜索

基于Faster R-CNN金丝猴优化检测方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索