Feature Data Measuring And Extracting Of Marine Plankton Using Digital Holographic Imaging

It is one important aspect of marine ecology research to study the advancedmethod of sensing marine plankton in order to effectively obtain critical marineplankton information data including species composition, their distribution,abundance and particle size, which contributes to the marine ecology research.Marine ecology researchers design some effective methods for sensing marineplankton to support to ocean ecology research. The goal of this paper is to use digitalholographic imaging for sensing marine plankton to recording sampling volume withseveral advantages such as no intrusive and nondestructive interrogation of therecording sampling volume. By capturing hologram of marine plankton andreconstructing hologram, the recorded optical field of marine plankton is retrieved,which provided the original image of marine plankton for further digital imageprocessing operations including filtering, binarization, canny operator edgeextraction, closing operation processing to obtain data of the particle field of themarina plankton. The biggest advantage of this method is able to achievenon-intrusive situ measurement, which can greatly reduce the disturbance to themarine plankton particle field and obtain the real data.In order to achieve this goal, first in this paper a brief review on the marineplankton observation research both at home and abroad is made. It studies the basictheory of digital holographic imaging for marine plankton from digital holographyrecording and reproduction and so on and discusses some structure parameters ofholographic recording devices and factors influencing the image reconstruction andultimately make sure of the use of coaxial holographic optical path; Second, two kindsof holographic reconstruction algorithms are studied and compared as well as thevalidation of digital holographic imaging method for marine plankton throughexperiments; Then problems that suppress the DC and conjugate image in thehologram are discussed. In this paper the quality of reconstructed image was obviously improved through removing the DC with the aid of the difference betweenhologram and the average intensity; Finally according to the characteristics of marineplankton, digital image processing technology is employed for the measurement andextraction of relevant characteristic parameter including the area, perimeter,equivalent diameter, round degree, compactness, centroid, transverse size,longitudinal dimension, shape ratio, seven invariant moments and ect., from whichdata with stability have been validated and extracted with aid of experiments.