图像自动识别技术在海洋浮游生物分析中的应用

介绍了一种利用数学形态学特征和Gabor纹理特征,结合主成分分析与支持向量机对胶州湾沿岸7种浮游生物的活体图像进行自动识别的方法。实验结果表明,基于主成分分析的降维识别模式可以提高系统识别性能,其平均识别正确率达78.5%,通过对图像采集、图像处理、特征的选取等方面做进一步的改进和提高,基于计算机数字图像的海洋浮游生物自动识别方法将为海洋生态环境监测提供新的实时、快速、高效检测平台。

Applications of automatic image identification for marine plankton analysis

In this paper,a method for the auto-identification of plankton images based on the combination of Principle Component Analysis and Support Vector Machine for contoured shape and Gabor texture classification is proposed.Seven species of algae and planktons sampled around Jiaozhou Bay were tested and it is shown that an overall 78.5% of classification accuracy is achieved with a reduced feature set based on Principle Component Analysis.The results showed that the image process system has potential to act as an efficient approach to mapping plankton populations in real time at open sea.