白蚁分区式微方向感知的精确形态识别

Precise morphological recognition with zonal micro-direction for termites

摘要

提出分区式微方向感知的自动形态分析方法,采用分区式思想细化目标,间接增加形态差异的显著性。选择旋转矩形表征目标和更多方向信息,在识别时引入多层局部空间感知模块,将方向与特征直接关联,同时融合双支空间金字塔模块,强化浅层特征复用,提高了计算效率。将主流旋转目标检测方法、关键点检测方法与本文方法对比,结果证明在提取高干扰的小目标的方向和位置方面,本文方法具有更好的精确度和鲁棒性。

Abstract

The partition-based approach is used in the paper to refine the target by indirectly enhancing the salience of morphological differences. Rotational moment-based representation is provided more directional information, and a multi-layer local spatial perception module is incorporated to directly associate direction with features. Furthermore, a dual-branch spatial pyramid module is introduced to enhance the reuse of shallow features and improve computational efficiency. In our experiments, the rotational object detection method, the keypoint detection method, and the proposed method are compared, and it is demonstrated that our method achieves better accuracy and robustness in extracting the direction and position of small targets under higher interference.

关键词

白蚁形态测量 / 旋转目标检测 / 关键点检测 / 物种鉴定 / 自动形态识别

Key words

measurement of termite morphology / oriented object detection / keypoint detection / species identification / automatic morphology recognition

参考文献

[1] 黄复生, 朱世模, 平正明, 等. 中国动物志(昆虫纲)[M]. 北京: 科学出版社, 2000: 1-961.

[2] HUANG Fusheng, ZHU Shimo, PING Zhengming, et al. Fauna sinica (insecta)[M]. Beijing: Science Press, 2000: 1-961.

[3] 白明, 杨星科. 三维几何形态学概述及其在昆虫学中的应用[J]. 昆虫学报, 2014, 57(9): 1105-1111.

[4] BAI Ming, YANG Xingke. A review of three—dimensional (3D) geo—metric morphometrics and its application in entomology[J]. Acta Entomologica Sinica, 2014, 57(9): 1105-1111.

[5] HOANG H N, HAI B H, PHUONG H L, et al. A lightweight keypoint matching framework for insect wing morphometric landmark detection[J]. Ecological Informatics, 2022, 70: 101694.

[6] JOÃO P R, WALTER G, ALICE M P. Automatic wing geometric morphometrics classification of honeybee (apis mellifera) subspecies using deep learning for detecting land—marks[J]. Big Data and Cognitive Computing, 2022, 6(3): 70— 70.

[7] WANG Zejun, ZHANG Shihao, CHEN Lijiao, et al. Microscopic insect pest detection in tea plantations: improved YOLOv8 model based on deep learning[J]. Agriculture, 2024, 14(10): 1739.

[8] BERECIARTUA—PÉREZ A, GOMEZ L, PICÓN A, et al. Insect counting through deep learning—based density map estimation[J]. Computers and Electronics in Agriculture, 2022, 197: 106933.

[9] BUSCHBACHER K, AHRENS D, ESPELAND M, et al. Image—based species identification of wild bees using convolutional neural networks[J]. Ecological Informatics, 2020, 55: 101017.

[10] BADGUJAR C M, ARMSTRONG P R, GERKEN A R, et al. Identifying common stored product insects using automated deep learning methods[J]. Journal of Stored Products Research, 2023, 103: 102166.

[11] VENEGAS P, CALDERON F, RIOFRÍO D, et al. Automatic ladybird beetle detection using deep—learning models[J]. PLoS One, 2021, 16(6): e0253027.

[12] TETILA E C, MACHADO B B, MENEZES G V, et al. A deep—learning approach for automatic counting of soybean insect pests[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 17(10): 1837-1841.

[13] 唐灿, 唐亮贵, 刘波. 图像特征检测与匹配方法研究综述[J]. 南京信息工程大学学报(自然科学版), 2020, 12(3): 261-273.

[14] TANG Can, TANG Lianggui, LIU Bo. A survey of image feature detection and matching methods[J]. Journal of Nanjing University of Information Science and Technology (Natural Science Edition), 2020, 12(3): 261-273.

[15] 江铁, 朱桂斌, 孙奥, 等. 特征点提取算法性能分析研究[J]. 科学技术与工程, 2012, 12(30): 7924-7930.

[16] JIANG Tie, ZHU Guibin, SUN Ao, et al. Performance analysis of feature point detectors[J]. Science Technology and Engineering, 2012, 12(30): 7924-7930.

[17] YANG Jieren, ZHANG xinying, LUO Peng. A summary of feature point detection based on image mosaics[J]. Electronic Test, 2021(6): 53-54.

[18] LE V, BEURTON—AIMAR M, ZEMMARI A, et al. Automated land—marking for insects morphometric analysis using deep neural networks[J]. Ecological Informatics, 2020, 60: 101175.

[19] GELDENHUYS D S, JOSIAS S, BRINK W, et al. Deep learning approaches to landmark detection in tsetse wing images[J]. PLOS Computational Biology, 2023, 19(6): e1011194.

[20] 安胜彪, 娄慧儒, 陈书旺, 等. 基于深度学习的旋转目标检测方法研究进展[J]. 电子测量技术, 2021, 44(21): 168-178.

[21] AN Shengbiao, LOU Huiru, CHEN Shuwang, et al. Research progress of rotating target detection methods based on deep learning[J]. Electronic Measurement Technology, 2021, 44(21): 168-178.

[22] FELZENSZWALB P F, GIRSHICK R B, MCALLESTER D, et al. Object detection with discriminatively trained part—based models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(9): 1627-1645.

[23] 陈天鹏, 胡建文. 基于深度学习的遥感图像旋转目标检测研究综述[J]. 计算机应用研究, 2024, 41(2): 329-340.

[24] CHEN Tianpeng, HU Jianwen. Overview of oriented object detection based on deep learning in remote sensing[J]. Application Research of Computers, 2024, 41(2): 329-340.

[25] LONG Wen, YU Cheng, YI Fang, et al. A comprehensive survey of oriented object detection in remote sensing images[J]. Expert Systems with Applications, 2023, 224: 16-26.

[26] 王旭, 吴艳霞, 张雪, 等. 计算机视觉下的旋转目标检测研究综述[J]. 计算机科学, 2023, 50(8): 79-92.

[27] WANG Xiu, WU Yanxia, ZHANG Xue, et al. Survey of rotating object detection research in computer vision[J]. Computer Science, 2023, 50(8): 79-92.

[28] YUAN Shuai, QIN Hanlin, YAN Xiang, et al. Sctransnet: spatial—channel cross transformer network for infrared small target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62(10): 1-15.

[29] YANG Zhigang, XIA Xiangyu, LIU Yiming, et al. LPST—Det: local—perception—enhanced swin transformer for sar ship detection[J]. Remote Sensing, 2024, 16(3): 18.

[30] KONG Tao, SUN Fuchun, LIU Huaping, et al. Foveabox: beyond anchor—based object detection[J]. IEEE Transactions on Image Processing, 2020, 29: 7389-7398.

[31] QIAN Wen, YANG Xue, PENG Silong, et al. Learning modulated loss for rotated object detection[EB/ OL]. (2019—11—09) [2025—08—15]. https:∥doi.org/ 10.48550/ arXiv.1911.08299.

[32] HAN Jiaming, DING Jian, LI Jie, et al. Align deep features for oriented object detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 10(99): 1-11.

[33] 中华人民共和国海关总署. SN/ T 1105—2019 大家白蚁检疫鉴定方法[S]. 北京: 中华人民共和国海关总署, 2019.

[34] General Administration of Customs of the People̓s Republic of China. SN/ T 1105—2019 Detection and identification of coptotermes curvignathus Holmgren[S]. Beijing: General Administration of Customs of the People̓s Republic of China, 2019.

[35] HOU Qibin, ZHOU Daquan, FENG Jiashi. Coordinate attention for efficient mobile network design[C]∥ Proceedings of the IEEE/ CVF Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE, 2021: 13708-13717.

[36] WANG Chenyao, LIAO Hongyuan, WU Yuehua, et al. CSPNET: a new backbone that can enhance learning capability of CNN[C]∥ Proceedings of the IEEE/ CVF Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE, 2020: 390-404.

基金资助

国家自然科学基金资助项目(62172095)

福建省自然科学基金资助项目(2025J01670)

福建省自然科学基金资助项目(2023J01515)