Abstract
Despite the rapid increase in the number and applications of plankton imaging systems in marine science, processing large numbers of images remains a major challenge due to large variations in image content and quality in different marine environments. We constructed an automatic plankton image recognition and enumeration system using an enhanced Convolutional Neural Network (CNN) and examined the performance of different network structures on automatic plankton image classification. The procedure started with an adaptive thresholding approach to extract Region of Interest (ROIs) from in situ plankton images, followed by a procedure to suppress the background noise and enhance target features for each extracted ROI. The enhanced ROIs were classified into seven categories by a pre-trained classifier which was a combination of a CNN and a Support Vector Machine (SVM). The CNN was selected to improve feature description and the SVM was utilized to improve classification accuracy. A series of comparison experiments were then conducted to test the effectiveness of the pre-trained classifier including the combination of CNN and SVM versus CNN alone, and the performance of different CNN models. Compared to CNN model alone, the combination of CNN and SVM increased classification accuracy and recall rate by 7.13% and 6.41%, respectively. Among the selected CNN models, the ResNet50 performed the best with accuracy and recall at 94.52% and 94.13% respectively. The present study demonstrates that deep learning technique can improve plankton image recognition and that the results can provide useful information on the selection of different CNN models for plankton recognition. The proposed algorithm could be generally applied to images acquired from different imaging systems.
Highlights
Zooplankton play a pivotal role in marine ecosystems by feeding on phytoplankton and serving as important food for fish larvae [1]
The first experiment (Table 1) was designed to examine whether incorporate the multiclass Support Vector Machine (SVM) would improve the performance of the Convolutional Neural Network (CNN) models
The baseline based on Histogram of Oriented Gradient (HOG) features and SVM classifier had relatively low performance on in situ plankton Region of Interest (ROIs) with precision and recall rates of approximately 60%
Summary
Zooplankton play a pivotal role in marine ecosystems by feeding on phytoplankton and serving as important food for fish larvae [1]. With the recent developments in machine learning, more sophisticated approaches such as Artificial Neural Networks (ANN) [26], random forest classifiers [27], Bayesian approaches [28], and Support Vector Machines (SVM) [9, 29, 30] have been applied to plankton classification Almost all these existing methods are customized descriptors that achieve the invariance by pre-selected rules and are not flexible enough to accommodate large variations in image quality and content in plankton images, e.g., morphological variation in the target objects caused by non-uniform illumination. Convergence speed and identification accuracy were increased by introducing shortcut connections between parameter layers Another recent development in CNN models is the Dense Convolutional Network (DenseNet) [37], which connects each layer to every other layer in a feed-forward fashion. The introduction of shortcut guarantees that the model makes full use of network residual information, which makes the topology of the network more complex, and improves the performance of the model with a much deeper layers
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
