基于人工智能的珊瑚物种歧视:Siderastrea大西洋综合体的案例研究
Marcos Soares Barbeitos1, Flávio Alberto Pérez1, Julián Olaya-Restrepo1
1Laboratório de Evolução dos Organismos Marinhos, Departamento de Zoologia, Universidade Federal do Paraná, Curitiba, Brazil.
PloS one
|December 11, 2024
概括
使用计算机视觉和人工智能在硬珊瑚中自动识别物种可以克服分类学挑战. 这种方法可以准确地区分物种,帮助研究人员进行硬珊瑚分类.
科学领域:
- 海洋生物学 海洋生物学
- 纳税学是一种分类学.
- 计算机科学 计算机科学
背景情况:
- 在Scleractinia (硬珊瑚) 中,由于定量特征和形态可塑性,物种划分是复杂的.
- 传统的分类学方法很难在硬珊瑚中定义清晰的物种界限.
- 人类对珊瑚形态物种的视觉歧视很难转化为客观描述.
研究的目的:
- 引入使用计算机视觉进行物种划界的形态特征的自动量化.
- 测试计算机视觉和人工智能的效率在有问题的Siderastrea属上.
- 开发一种能够模拟人类对硬珊瑚分类学歧视能力的工具.
主要方法:
- 利用完成的本地二进制模式 (CLBP) 来自动量化形态特征.
- 使用带有模糊逻辑的人工神经网络 (Θ-FAM) 进行监督分类,处理不确定性.
- 将CLBP和 Θ-FAM 的性能与传统的形态测量和线性差异分析 (LDA) 的性能进行比较.
主要成果:
- 使用CLBP和 Θ-FAM 的自动化物种识别显著超过了与 Θ-FAM 结合的传统形态学字符.
- 此外,CLBP和Θ-FAM方法也优于CLBP与LDA结合的方法.
- 机器学习统计证实了人工智能驱动方法的有效性.
结论:
- 计算机视觉和人工智能可以在硬珊瑚分类学中成功模拟人类视觉歧视能力.
- 这种自动化方法为克服Scleractinia的分类学障碍提供了一个有价值的工具.
- 这些发现支持使用人工智能来解决海洋生物中复杂的物种划分问题.
相关概念视频
01:23Red Algae
Red algae, also known as rhodophytes, are primarily found in marine environments, though some species inhabit freshwater and terrestrial ecosystems. These organisms exist in both unicellular and multicellular forms, with some multicellular varieties reaching macroscopic sizes.As phototrophic organisms, red algae contain chlorophyll a; however, their chloroplasts lack chlorophyll b. Instead, they possess phycobiliproteins, which serve as major light-harvesting pigments, similar to those found in...
01:46Types of Selection
40.2K
Natural selection influences the frequencies of particular alleles and phenotypes within populations in several different ways. Primarily, natural selection can be directional, stabilizing, or disruptive. Directional selection favors one extreme trait and shifts the population towards that phenotype while selecting against individuals displaying alternate traits. Stabilizing selection favors an intermediate trait with a narrow range of variation. Deviation from the optimal phenotype towards an...
40.2K