项目名称: 基于生物形态特征的中国海常见有害赤潮藻显微图像识别
项目编号: No.61271406
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 无线电电子学、电信技术
项目作者: 姬光荣
作者单位: 中国海洋大学
项目金额: 85万元
中文摘要: 中国海常见有害赤潮藻的自动识别对我国海洋生态系统、环境监测和海洋渔业生产等多方面具有重要的科学和现实意义。传统的赤潮藻鉴定主要依赖经验丰富的藻类专家依据种的生物形态学特征在显微镜下通过人工目视判读、分类,该方法存在专业水平高、分类人员断层、耗时费力等问题。而目前国内外基于计算机视觉的藻细胞显微图像自动识别研究存在难以有效地分割图像目标、提取表示特征等问题。本项目针对41种中国海常见有害赤潮藻,提出依据藻种的角毛、横纵沟、尖顶刺等生物形态特征建立藻种表示特征集合;根据生物学分类方法建立分类器,实现中国海常见有害赤潮藻的自动分类。 此项研究结合海洋生物学家的藻种分类知识,综合运用图像处理、分析、模式识别技术,建立赤潮藻自动识别系统,有望实现赤潮生物优势藻种快速、有效的鉴定与检测,为赤潮现场监测系统的研究奠定基础。
中文关键词: 层级分类器;图像分割;特征提取;图像分析;图像识别
英文摘要: Automatic identification of Harmful Algal Blooms (HABs) in Chinese coast areas has a very important scientific and practical significance for our marine ecosystems, environmental monitoring and marine fishery. Traditionally identification of HABs mainly relies on well-experienced algae operators using microscope based on characteristics of biological morphology, which need high level professional experts that are becoming less and less. Besides, it is also time-consuming and laborious. However, the current research on automatic identification and classification of phytoplankton by microscopic images based on computer vision technology is difficult to segment cell target and extract characteristics effectively. The proposed research focuses on 41 kinds of HABs in Chinese coast areas and proposes to create the feature sets of phytoplankton according to their biomorphic characteristics such as seta, cingulum and sulcus, and spine; constructs classifier according to biological classification method to achieve the automatic identification of HABs in Chinese coast areas. The proposed research combines the knowledge of phytoplankton classification by biologist with image processing, image analysis as well as pattern recognition technology to establish the HAB automatic identification system, in order to realize the rap
英文关键词: hierarchical classifier;image segmentation;feature extraction;image analysis;image recognition