中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 合肥物质科学研究院
  3. 中国科学院合肥物质科学研究院
Applying feature-similarity-metrics for long-tailed problem of phytoplankton microscopic images classification

文献类型：期刊论文

作者Liang, Tianhong1,2; Yin, Gaofang1,2,3; Zhao, Nanjing1,2,3; Jia, Renqing1; Zhang, Xiaoling4; Xu, Min1,2; Zhang, Zihao4; Dong, Ming3; Hu, Xiang1; Huang, Peng1
刊名ALGAL RESEARCH-BIOMASS BIOFUELS AND BIOPRODUCTS
出版日期2024-08-01
卷号82
关键词Phytoplankton Microscopic image Deep learning Feature-similarity-metrics Feature-center-constraint Long-tailed distribution
ISSN号2211-9264
DOI10.1016/j.algal.2024.103673
通讯作者Yin, Gaofang(gfyin@aiofm.ac.cn) ; Zhao, Nanjing(njzhao@aiofm.ac.cn)
英文摘要The distribution of phytoplankton in natural water bodies holds significant importance for aquatic for maintaining healthy aquatic environments. While deep learning has become a popular research field of automating phytoplankton identification, its performance is diminished by the uneven species distribution in nature world, which biases classification models towards more advantaged species. Addressing this long-tailed problem, the paper proposed a novel deep learning idea termed feature-similarity-metrics species which utilizes the principle of decreasing intra-class differences and increasing the inter-class differences. A method called feature-center-constraint is introduced to implement the idea. Initially, a center is presupposed for each category, and these centers are distributed evenly and sparsely within the feature space using a uniform distribution initialization method. During model training, the Euclidean distance is used to measure the similarity between sample features and corresponding center. This approach enhanced the model's ability to predict disadvantaged phytoplankton species without compromising the classification accuracy of advantaged ones. Using a phytoplankton dataset of 26,564 images from 68 genera, collected from Chaohu Lake, this method demonstrates superior performance compared to existing techniques. Specially, it achieves a 5.69 % increase in micro-average F1 score and an 11.85 % rise in macro-average F1 score over general method, while F1 score is widely recognized as a reliable metric for measuring classification models' performance. These enhancements not only increase the accuracy of automated phytoplankton identification but also facilitate more effective monitoring of biodiversity and ecological health in aquatic systems. This makes the research highly beneficial for practical applications in environmental science, offering rapid and accurate insights into phytoplankton community structures, for ecological assessments.
WOS关键词MICROALGAE
资助项目Major Science and Technology Projects in Anhui Province[202203a07020002] ; Institute of environment, Hefei Comprehensive National Science Center Research Team Construction Project[HYKYTD2024004] ; Anhui Provincial Ecological Environment Research Project[2023hb0011] ; Anhui Provincial Major Science and Technology Projects Technology Innovation Platform[S202305a12020004] ; HFIPS Director's Fund[YZJJ2024QN01]
WOS研究方向Biotechnology & Applied Microbiology
语种英语
WOS记录号WOS:001316470100001
出版者ELSEVIER
资助机构Major Science and Technology Projects in Anhui Province ; Institute of environment, Hefei Comprehensive National Science Center Research Team Construction Project ; Anhui Provincial Ecological Environment Research Project ; Anhui Provincial Major Science and Technology Projects Technology Innovation Platform ; HFIPS Director's Fund
源URL[http://ir.hfcas.ac.cn:8080/handle/334002/135464]  
专题中国科学院合肥物质科学研究院
通讯作者Yin, Gaofang; Zhao, Nanjing
作者单位1.Chinese Acad Sci, Anhui Inst Opt & Fine Mech, Hefei 230031, Peoples R China
2.Univ Sci & Technol China, Hefei 230026, Peoples R China
3.Hefei Comprehens Natl Sci Ctr, Inst Environm, Hefei 230031, Peoples R China
4.Anhui Univ, Hefei 230039, Peoples R China
推荐引用方式
GB/T 7714		 Liang, Tianhong,Yin, Gaofang,Zhao, Nanjing,et al. Applying feature-similarity-metrics for long-tailed problem of phytoplankton microscopic images classification[J]. ALGAL RESEARCH-BIOMASS BIOFUELS AND BIOPRODUCTS,2024,82.
APA Liang, Tianhong.,Yin, Gaofang.,Zhao, Nanjing.,Jia, Renqing.,Zhang, Xiaoling.,...&Huang, Peng.(2024).Applying feature-similarity-metrics for long-tailed problem of phytoplankton microscopic images classification.ALGAL RESEARCH-BIOMASS BIOFUELS AND BIOPRODUCTS,82.
MLA Liang, Tianhong,et al."Applying feature-similarity-metrics for long-tailed problem of phytoplankton microscopic images classification".ALGAL RESEARCH-BIOMASS BIOFUELS AND BIOPRODUCTS 82(2024).

入库方式： OAI收割

来源：合肥物质科学研究院

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。