Classification of marine microalgae using low-resolution Mueller matrix images and convolutional neural network

Abstract

In this paper, we used a convolutional neural network to study the classification of marine microalgae by using low-resolution Mueller matrix images. Mueller matrix images of 12 species of algae from 5 families were measured by a Mueller matrix microscopy with an LED light source at 514 nm wavelength. The data sets of seven resolution levels were generated by the bicubic interpolation algorithm. We conducted two groups of classification experiments; one group classified the algae into 12 classes according to species category, and the other group classified the algae into 5 classes according to family category. In each group of classification experiments, we compared the classification results of the Mueller matrix images with those of the first element (M11) images. The classification accuracy of Mueller matrix images declines gently with the decrease of image resolution, while the accuracy of M11 images declines sharply. The classification accuracy of Mueller matrix images is higher than that of M11 images at each resolution level. At the lowest resolution level, the accuracy of 12-class classification and 5-class classification of full Mueller matrix images is 29.89% and 35.83% higher than those of M11 images, respectively. In addition, we also found that the polarization information of different species had different contributions to the classification. These results show that the polarization information can greatly improve the classification accuracy of low-resolution microalgal images.

Detection Methods of Nanoparticles Synthesized by Gas-Phase Method: A Review

“…At the same time, the polarized light path need not be designed separately, as long as the current dynamic light scattering light path-polarizing element can be added. At present, it has been applied to biomedicine ( Xiang-Yu et al, 2019 ; He et al, 2021 ; Meng et al, 2021 ; Shen et al, 2021 ; Song et al, 2021 ), marine microbial detection ( Wang et al, 2018 ; Li et al, 2020; Liu et al, 2020 ; Wang et al, 2020 ; Li et al, 2021 ), atmospheric detection ( Li et al, 2019 ; Qizhi et al, 2021 ), etc. It is believed that with the introduction of polarized light technology, the accuracy, measurable parameters, and applicable range of the dynamic light scattering method will be greatly improved.…”

“…Therefore, the development of corresponding detection technology is of great significance to evaluate the properties of nanomaterials and improve the preparation efficiency and quality of nanoparticles. At present, the detection technology mainly focuses on the measurement of macroscopic statistical characteristics and microstructure characteristics of nanoparticles (Liu et al, 2005). The commonly used detection methods can be divided into two types according to the detection principle: (1) the optical-based detection method; (2) the electrical-based detection method; and (3) the mechanical-based detection method.…”

A Method for Optimal Detection of Lung Cancer Based on Deep Learning Optimized by Marine Predators Algorithm

“…If the attraction sense for the prey in the hunting region is high, predators employ Brownian method; if not so, if attraction sense of prey is low, Lévy method has been used. Ecological problems such as fish aggregating devices (FADs) effects or eddy formation alternates the marine predator's behavior [19]. In MPA, at first, initial n numbers of population for the prey are defined with considering a lower and an upper bound, i.e.,…”

Combined Mueller matrix imaging and artificial intelligence classification framework for Hepatitis B detection

“…Liu et al. 28 classified marine microalgae using a low-resolution MMIP technique and a CNN and showed that the classification accuracy obtained using the whole Mueller matrix image was greater than that achieved using the image alone at each resolution level. Ma et al.…”