Phase congruency-based detection of circular objects applied to analysis of phytoplankton images

Verikas, Antanas; Gelžinis, Adas; Bačauskienė, Marija; Olenina, Irina; Olenin, Sergej; Vaičiukynas, Evaldas

doi:10.1016/j.patcog.2011.10.019

Cited by 47 publications

(33 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3, majority of small "uninteresting" objects were eliminated and all P. minimum cells, except the two marked by arrows and those with center points too close to the image boundaries, were detected. Note that objects with center points too close to image boundaries were left out of the analysis, due to constraints imposed by the phase-congruency-based image preprocessing [36]. When using fluorescence, all erroneous detections due to small "uninteresting" objects [clearly seen in the upper left and bottom right parts of the image shown in Fig.…”

Section: B Cell Detection Resultsmentioning

confidence: 99%

“…1) object geometry; 2) Fourier descriptors; 3) contour curvature; 4) image properties near the contour; 5) cell image properties. Compared to our previous work [36], the feature set used in this study was extended by groups 2), 3), and 4), and by adding several features to group 1).…”

Section: B Feature Extractionmentioning

confidence: 99%

“…Each coordinate pair can be treated as a complex number so that (4) The discrete Fourier transform of is Table II. 5) Cell Image Properties: A variety of features are used to characterize cell image properties. Among others, we also use features computed from the so-called , and images obtained by applying the phase-congruency-based image preprocessing [35], [36]. A complete list of features used to characterize cell image properties is given in Table III. IV.…”

Section: ) Fourier Descriptorsmentioning

confidence: 99%

“…We used the same genetic algorithm as in [36], to select features for SVM and to determine the optimal values of SVM hyperparameters.…”

Section: Feature Selectionmentioning

confidence: 99%

“…The fluorescence microscopy is of great help in finding light microscopy image areas containing P. minimum cells, thus, mitigating the object detection problem. Various image preprocessing techniques, including phase-congruency-based image preprocessing [35], [36], are applied to extract a variety of features characterizing cells and cell contours in light microscopy images. Relevant feature subsets are then selected and used in SVM as well as RF classifiers to distinguish between P. minimum cells and other objects.…”

mentioning

confidence: 99%

See 4 more Smart Citations

An Integrated Approach to Analysis of Phytoplankton Images

Verikas

Gelžinis

Bačauskienė

et al. 2015

IEEE J. Oceanic Eng.

View full text Add to dashboard Cite

Abstract-The main objective of this paper is detection, recognition, and abundance estimation of objects representing the Prorocentrum minimum (Pavillard) Schiller (P. minimum) species in phytoplankton images. The species is known to cause harmful blooms in many estuarine and coastal environments. The proposed technique for solving the task exploits images of two types, namely, obtained using light and fluorescence microscopy. Various image preprocessing techniques are applied to extract a variety of features characterizing P. minimum cells and cell contours. Relevant feature subsets are then selected and used in support vector machine (SVM) as well as random forest (RF) classifiers to distinguish between P. minimum cells and other objects. To improve the cell abundance estimation accuracy, classification results are corrected based on probabilities of interclass misclassification. The developed algorithms were tested using 158 phytoplankton images. There were 920 P. minimum cells in the images in total. The algorithms detected 98.1% of P. minimum cells present in the images and correctly classified 98.09% of all detected objects. The classification accuracy of detected P. minimum cells was equal to 98.9%, yielding a 97.0% overall recognition rate of P. minimum cells. The feature set used in this work has shown considerable tolerance to out-of-focus distortions. Tests of the system by phytoplankton experts in the cell abundance estimation task of P. minimum species have shown that its performance is comparable or even better than performance of phytoplankton experts exhibited in manual counting of artificial microparticles, similar to P. minimum cells. The automated system detected and correctly recognized 308 (91.1%) of 338 P. minimum cells found by experts in 65 phytoplankton images taken from new phytoplankton samples and erroneously assigned to the P. minimum class 3% of other objects. Note that, due to large variations of texture and size of P. minimum cells as well as background, the task performed by the system was more complex than that performed by the experts when counting artificial microparticles similar to P. minimum cells.

show abstract

Section: B Cell Detection Resultsmentioning

confidence: 99%

Section: B Feature Extractionmentioning

confidence: 99%

Section: ) Fourier Descriptorsmentioning

confidence: 99%

“…We used the same genetic algorithm as in [36], to select features for SVM and to determine the optimal values of SVM hyperparameters.…”

Section: Feature Selectionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

An Integrated Approach to Analysis of Phytoplankton Images

Verikas

Gelžinis

Bačauskienė

et al. 2015

IEEE J. Oceanic Eng.

View full text Add to dashboard Cite

show abstract

Automatic setae segmentation fromChaetocerosmicroscopic images

Zheng

Zhao

Sun

et al. 2014

Microsc. Res. Tech.

View full text Add to dashboard Cite

A novel image processing model Grayscale Surface Direction Angle Model (GSDAM) is presented and the algorithm based on GSDAM is developed to segment setae from Chaetoceros microscopic images. The proposed model combines the setae characteristics of the microscopic images with the spatial analysis of image grayscale surface to detect and segment the direction thin and long setae from the low contrast background as well as noise which may make the commonly used segmentation methods invalid. The experimental results show that our algorithm based on GSDAM outperforms the boundary-based and region-based segmentation methods Canny edge detector, iterative threshold selection, Otsu's thresholding, minimum error thresholding, K-means clustering, and marker-controlled watershed on the setae segmentation more accurately and completely.

show abstract