In-situ classification and identification method and device of marine phytoplankton by microscopic fluorescence spectral imaging

Abstract

The invention discloses an in-situ classification and identification method of marine phytoplankton by microscopic fluorescence spectral imaging, and further discloses an in-situ classification and identification device of the marine phytoplankton by the microscopic fluorescence spectral imaging, and aims at combining a fluorescence microscopy with spectral imaging techniques to develop rapid microscopic fluorescence spectrum detection techniques which can be applied to algae in marine environments, obtain image arrays of similar algae at different wavelengths, take three-dimensional coordinate values of spectrums embodied in colorimetry under different excitation wavelengths as additional information of algal images, and extract features from different images of a same species obtained under the different wavelengths and train a classifier to realize classification and identification and statistics of algae species with similar shapes inside a same class group. An imaging system is divided into a laboratory scheme and an underwater scheme. The two sets of systems take actual requirements which can be complementary and verified to each other into consideration under different conditions and at different precisions.

Description

technical field [0001] The invention relates to a method and a device for in-situ classification and identification of marine phytoplankton by microscopic fluorescence spectrum imaging. Background technique [0002] The monitoring of species and quantity of phytoplankton is a routine basic work in marine scientific research. The exploration of fast, accurate and in-situ automatic identification and analysis methods of phytoplankton and the establishment of the system have important theoretical and practical significance; it can not only enrich the taxonomic research of phytoplankton, but also realize the change, occurrence, Real-time detection of development and extinction process, and can be widely used in marine ecological survey, environmental monitoring, aquaculture, red tide, geology and other fields. This is the common goal pursued by domestic and foreign counterparts. [0003] Among the various automatic analysis and identification methods of phytoplankton that have...

AI Technical Summary

Problems solved by technology

With the rapid development of various image feature extraction and classifier methods, the research on methods and algorithms for automatic classification and recognition of different forms of algae is a common research idea at present for the image information obtained by the microscope. The image recognition method based on the excitation of a single light source is still difficult to identify species with similar shapes within the same group, and automatic image classification and recognition is even more difficult to achieve

Images