Fully automatic detection and classification of phytoplankton specimens in digital microscopy images

BACKGROUND AND OBJECTIVE The proliferation of toxin-producing phytoplankton species can compromise the quality of the water sources. This contamination is difficult to detect, and consequently to be neutralised, since normal water purification techniques are ineffective. Currently, the water analyses about phytoplankton are commonly performed by the specialists with manual routine analyses, which represents a major limitation. The adequate identification and classification of phytoplankton specimens requires intensive training and expertise. Additionally, the performed analysis involves a lengthy process that exhibits serious problems of reliability and repeatability as inter-expert agreement is not always reached. Considering all those factors, the automatization of these analyses is, therefore, highly desirable to reduce the workload of the specialists and facilitate the process. METHODS This manuscript proposes a novel fully automatic methodology to perform phytoplankton analyses in digital microscopy images of water samples taken with a regular light microscope. In particular, we propose a method capable of analysing multi-specimen images acquired using a simplified systematic protocol. In contrast with prior approaches, this enables its use without the necessity of an expert taxonomist operating the microscope. The system is able to detect and segment the different existing phytoplankton specimens, with high variability in terms of visual appearances, and to merge them into colonies and sparse specimens when necessary. Moreover, the system is capable of differentiating them from other similar objects like zooplankton, detritus or mineral particles, among others, and then classify the specimens into defined target species of interest using a machine learning-based approach. RESULTS The proposed system provided satisfactory and accurate results in every step. The detection step provided a FNR of 0.4%. Phytoplankton detection, that is, differentiating true phytoplankton from similar objects (zooplankton, minerals, etc.), provided a result of 84.07% of precision at 90% of recall. The target species classification, reported an overall accuracy of 87.50%. The recall levels for each species are, 81.82% for W. naegeliana, 57.15% for A. spiroides, 85.71% for D. sociale and 95% for the "Other" group, a set of relevant toxic and interesting species widely spread over the samples. CONCLUSIONS The proposed methodology provided accurate results in all the designed steps given the complexity of the problem, particularly in terms of specimen identification, phytoplankton differentiation as well as the classification of the defined target species. Therefore, this fully automatic system represents a robust and consistent tool to aid the specialists in the analysis of the quality of the water sources and potability.

[1]  Hansang Lee,et al.  Plankton classification on imbalanced large scale database via convolutional neural networks with transfer learning , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[2]  Marc Acheroy,et al.  Texture classification using Gabor filters , 2002, Pattern Recognit. Lett..

[3]  P. Culverhouse,et al.  Do experts make mistakes? A comparison of human and machine identification of dinoflagellates , 2003 .

[4]  Peter Gais,et al.  Automatic analysis of aqueous specimens for phytoplankton structure recognition and population estimation , 2006, Microscopy research and technique.

[5]  Heidi M. Sosik,et al.  WHOI-Plankton- A Large Scale Fine Grained Visual Recognition Benchmark Dataset for Plankton Classification , 2015, ArXiv.

[6]  K. Vuorio,et al.  Intercalibrations of freshwater phytoplankton analyses , 2007 .

[7]  Silvia Silva da Costa Botelho,et al.  Deep Learning for Microalgae Classification , 2017, 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA).

[8]  Rita K. Henderson,et al.  A review of monitoring technologies for real-time management of cyanobacteria: Recent advances and future direction , 2016 .

[9]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[10]  H. Enevoldsen,et al.  Addressing the Problem of Harmful Algal Blooms in Latin America and the Caribbean- A Regional Network for Early Warning and Response , 2018, Front. Mar. Sci..

[11]  Qiao Hu,et al.  Automatic plankton image recognition with co-occurrence matrices and Support Vector Machine , 2005 .

[12]  Matthew R. First,et al.  Performance of the human “counting machine”: evaluation of manual microscopy for enumerating plankton , 2012 .

[13]  Stefan Poslad,et al.  An Enhanced Bag-of-Visual Word Vector Space Model to Represent Visual Content in Athletics Images , 2012, IEEE Transactions on Multimedia.

[14]  Eva Álvarez,et al.  Improvement of plankton biovolume estimates derived from image-based automatic sampling devices: application to FlowCAM , 2012 .

[15]  Robert J. Olson,et al.  An automated submersible flow cytometer for analyzing pico- and nanophytoplankton: FlowCytobot , 2003 .

[16]  Lynne Boddy,et al.  Identification of 72 phytoplankton species by radial basis function neural network analysis of flow cytometric data , 2000 .

[17]  P. Culverhouse,et al.  Automatic classification of field-collected dinoflagellates by artificial neural network , 1996 .

[18]  Lawrence O. Hall,et al.  Recognizing plankton images from the shadow image particle profiling evaluation recorder , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[19]  Oscar Beijbom,et al.  Transfer Learning and Deep Feature Extraction for Planktonic Image Data Sets , 2017, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV).

[20]  Matthew B. Blaschko,et al.  Combining Local and Global Image Features for Object Class Recognition , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[21]  Joaquim de Moura,et al.  Deep Feature Analysis in a Transfer Learning-based Approach for the Automatic Identification of Diabetic Macular Edema , 2019, 2019 International Joint Conference on Neural Networks (IJCNN).

[22]  Thomas Dandekar,et al.  PlanktoVision – an automated analysis system for the identification of phytoplankton , 2013, BMC Bioinformatics.

[23]  Paulo Drews,et al.  Supervised Microalgae Classification in Imbalanced Dataset , 2016, 2016 5th Brazilian Conference on Intelligent Systems (BRACIS).

[24]  Keiichi Abe,et al.  Topological structural analysis of digitized binary images by border following , 1985, Comput. Vis. Graph. Image Process..

[25]  K. V. Embleton,et al.  Automated counting of phytoplankton by pattern recognition: a comparison with a manual counting method , 2003 .

[26]  C. Davis.  The Video Plankton Recorder (VPR) : Design and initial results , 1992 .

[27]  Richard J. Telford,et al.  HUMAN ERROR AND QUALITY ASSURANCE IN DIATOM ANALYSIS , 2002 .

[28]  Nan Wang,et al.  Robust and automatic cell detection and segmentation from microscopic images of non-setae phytoplankton species , 2017, IET Image Process..

[29]  Antonio Criminisi,et al.  Object categorization by learned universal visual dictionary , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[30]  Antanas Verikas,et al.  Phase congruency-based detection of circular objects applied to analysis of phytoplankton images , 2012, Pattern Recognit..

[31]  Charles S. Yentsch,et al.  An imaging-in-flow system for automated analysis of marine microplankton , 1998 .

[32]  M. R. Turner,et al.  Texture discrimination by Gabor functions , 1986, Biological Cybernetics.

[33]  H. Paerl,et al.  Climate change: links to global expansion of harmful cyanobacteria. , 2012, Water research.

[34]  Antanas Verikas,et al.  A novel technique to extract accurate cell contours applied for segmentation of phytoplankton images , 2014, Machine Vision and Applications.

[35]  C. V. Jawahar,et al.  Word Image Retrieval Using Bag of Visual Words , 2012, 2012 10th IAPR International Workshop on Document Analysis Systems.

[36]  Antony M. Knights,et al.  From microscope to management: the critical value of plankton taxonomy to marine policy and biodiversity conservation , 2017 .

[37]  W. K. Stewart,et al.  Rapid visualization of plankton abundance and taxonomic composition using the Video Plankton Recorder , 1996 .

[38]  A. Solow,et al.  Microaggregations of Oceanic Plankton Observed by Towed Video Microscopy , 1992, Science.

[39]  Scott Samson,et al.  A system for high-resolution zooplankton imaging , 2001 .

[40]  Hayato Kondo,et al.  Development of a realtime plankton image archiver for AUVs , 2014, 2014 IEEE/OES Autonomous Underwater Vehicles (AUV).

[41]  Derek Bradley,et al.  Adaptive Thresholding using the Integral Image , 2007, J. Graph. Tools.

[42]  Juan José del Coz,et al.  Automatic plankton quantification using deep features , 2019, Journal of Plankton Research.

[43]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[44]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[45]  Y. Pachepsky,et al.  Identification and enumeration of cyanobacteria species using a deep neural network , 2020 .

[46]  Iraklis Varlamis,et al.  Use of health-related indices and classification methods in medical data , 2019, Classification Techniques for Medical Image Analysis and Computer Aided Diagnosis.

[47]  Hai Guo,et al.  A Research on the Recognition of Chironomid Larvae Based on SVM , 2009, 2009 Pacific-Asia Conference on Circuits, Communications and Systems.