基于成像模型和傅里叶描述符的微生物轮廓分割方法  

Microorganism Contour Segmentation Method Using Imaging Models and Fourier Descriptors

作　　者：杨昊 史圣兵 姚亮亮 桂甸 时璐 赵金宇[1] 孟浩然[1] Yang Hao;Shi Shengbing;Yao Liangliang;Gui Dian;Shi Lu;Zhao Jinyu;Meng Haoran(Key Laboratory of Advanced Manufacturing of Optical Systems,Changchun Institute of Optics,Precision Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,Jilin,China;University of Chinese Academy of Sciences,Beijing 100049,China;Chinese People’s Liberation Army Unit 63850,Baicheng 137000,Jilin,China;Army Equipment Department Military Representative Office in Chongqing Region,Kunming Branch,Kunming 650000,Yunnan,China)

机构地区：[1]中国科学院长春光学精密机械与物理研究所光学系统先进制造重点实验室,吉林长春130033 [2]中国科学院大学,北京100049 [3]解放军63850部队,吉林白城137000 [4]陆军装备部驻重庆地区军事代表局驻昆明代表室,云南昆明650000

出　　处：《激光与光电子学进展》2025年第2期378-389,共12页Laser & Optoelectronics Progress

基　　金：国家重点研发计划(2021YFC3101300)。

摘　　要：海洋微生物是海洋生态系统的基础,针对水下成像过程中因水体吸收散射引起的图像中微生物轮廓模糊问题,提出一种结合成像模型和傅里叶描述符的水下微生物轮廓分割方法。首先通过水下成像模型估计背景光和水体衰减系数,从而提取对象的无水特征图。然后通过分类头确定目标位置,回归头则基于傅里叶描述符获取微生物的轮廓表示,并在像素域中进一步细化轮廓。此外,还进行了全息图重构和预处理,并制备了微生物轮廓分割数据集。实验结果表明,傅里叶描述符相较于星状多边形方法,在轮廓表示精度和空间连续性上均具有优越性。在与传统分割方法的对比中,本文算法的F1分数为0.8894、交并比为0.7887、像素精确度为0.8608,相较其他算法均有提高,表明本文算法具有较好的分割能力。Marine microorganisms are fundamental to marine ecosystems.However,underwater imaging often blurs microbial contours due to water absorption and scattering.To address this,we propose a contour segmentation method for underwater microorganisms that combines an underwater imaging model with Fourier descriptors.First,the background light and water attenuation coefficients are estimated using the underwater imaging model to extract a clear,water-free feature map of the object.Next,a classification header determines the target location,while a regression header uses Fourier descriptors to represent and refine the microorganism’s contour in the pixel domain.In addition,hologram reconstruction and preprocessing steps are applied,and a microbial contour segmentation dataset is generated.Experimental results demonstrate that the Fourier descriptor outperforms the star polygon method in contour representation accuracy and spatial continuity.Compared to traditional segmentation methods,the proposed algorithm achieves an F1 score of 0.8894,intersection over union of 0.7887,and pixel accuracy of 0.8608,all improved metrics indicating superior segmentation capability.

关 键 词：实例分割 水下图像处理 水下成像模型 傅里叶描述符 

分 类 号：TP751.2[自动化与计算机技术—检测技术与自动化装置]