Autonomous agents for edge detection and continuity perception on otolith images

Guillaud, Anne; Benzinou, Abdesslam; Troadec, H.; Bihan, Jean Le

doi:10.1016/s0262-8856(02)00102-6

Cited by 16 publications

(36 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, as reported in Guillaud et al (2002), previously proposed 1D and 2D techniques are both outperformed by the proposed statistical learning scheme with, respectively, 50 and 80% of correct classification. In addition, compared to 2D approaches which require about 1 min per image, the SVM classification from PB features just needs a few seconds on a 2.4 GHz PC in a Matlab implementation.…”

Section: Extension To Image Classificationsupporting

confidence: 65%

“…The main drawback lies in the absence of 2D perception of ring continuity, being thus sensitive to local artifacts in otolith images. The evaluation carried out on plaice otolith sets shows that these 1D approaches achieve only low performance (Guillaud et al, 2002). On the other hand, 2D ring segmentation has also been investigated: deformable templates deduced from the external shape of the otolith (Traodec et al, 2000), a graph framework (Rodin et al, 1996), locally deformable spline-based models (Benzinou et al, 1997) and multi-agent systems (Guillaud et al, 2002).…”

Section: Problem Statement and Related Workmentioning

confidence: 99%

“…The evaluation carried out on plaice otolith sets shows that these 1D approaches achieve only low performance (Guillaud et al, 2002). On the other hand, 2D ring segmentation has also been investigated: deformable templates deduced from the external shape of the otolith (Traodec et al, 2000), a graph framework (Rodin et al, 1996), locally deformable spline-based models (Benzinou et al, 1997) and multi-agent systems (Guillaud et al, 2002). Significant improvements were obtained for fish age estimation (Guillaud et al, 2002) compared to 1D approaches.…”

Section: Problem Statement and Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Automated fish age estimation from otolith images using statistical learning

Fablet

Josse

2005

Fisheries Research

View full text Add to dashboard Cite

The acquisition of age and growth data is of key importance for fisheries research (assessment, marine ecology issues, etc.). Consequently, automating this task is of great interest. In this paper, we investigate the use of statistical learning techniques for fish age estimation. The core of this study lies in the definition of relevant image-related features. We rely on the computation of a 1D representation summing up the content of otolith images within a predefined area of interest. Features are then extracted from this non-stationary representation depicting the alternation of seasonal growth rings. Thus, fish age estimation can be viewed as a multi-class classification issue using statistical learning strategies. In particular, a procedure based on demodulation and remodulation of fish growth patterns is used to improve the generalization properties of the trained classifiers. The experimental evaluation is carried out over a dataset of 320 plaice otolith images from age groups 1-6. We analyze both, the performances of several statistical classifiers, namely SVMs (support vector machines) and neural networks, and the relevance of the proposed image-based feature sets. In addition, the combination of additional biological and shape features to the image-related ones is considered. We reach a rate of correct age estimation of 88% w.r.t. the expert ground truth. This demonstrates the relevance of the proposed approach for the automation of routine aging and for computer-assisted aging.

show abstract

Section: Extension To Image Classificationsupporting

confidence: 65%

Section: Problem Statement and Related Workmentioning

confidence: 99%

Section: Problem Statement and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Automated fish age estimation from otolith images using statistical learning

Fablet

Josse

2005

Fisheries Research

View full text Add to dashboard Cite

show abstract

“…Modification of intensities and multilevel thresholding have sometimes been effective in cases of inhomogeneity (Luo et al, 1998). In addition to thresholding, several methods have been suggested to automatically extract objects from images (Guillaud et al, 2002;Ji and Yan, 2002). However, fully automated and universal procedures are still unavailable.…”

Section: Comparisons Of Reant With Often-used 3-dmentioning

confidence: 99%

Morphometrical study of plant vacuolar dynamics in single cells using three-dimensional reconstruction from optical sections

Kutsuna

Hasezawa

2005

Microsc. Res. Tech.

View full text Add to dashboard Cite

In higher plants, vacuoles increase their volumes in accordance with cell enlargement and occupy most of the cell volume. However, quantitative analyses of vacuolar contributions during changes in cell morphology have been hampered by the inadequacies and frequent artifacts associated with current three-dimensional (3-D) reconstruction methods of images derived from light microscopy. To overcome the limitations of quantifying 3-D structures, we have introduced 3-D morphometrics into light microscopy, adopting a contour-based approach for which we have developed an interpolation method. Using this software, named REANT, the morphological and morphometrical changes in protoplasts and vacuoles during plasmolysis could be investigated. We employed the tobacco (Nicotiana tabacum) BY-2 cell line No.7, expressing a GFP-AtVam3p fusion protein, BY-GV7, using GFP as a marker of vacuolar membranes (VMs). By vital staining of the plasma membrane (PM) of cells, we simultaneously obtained optical sections of both the PM and VM. We, therefore, reconstructed the 3-D structures of protoplasts and vacuoles before and after plasmolysis. We were able to identify the appearance of elliptical structures of VMs in the vacuolar lumen, and to determine that they were derived from cytoplasmic strands. From the 3-D structures, the volumes and surface areas were measured at the single cell level. The shrinkage of vacuoles accounted for most of the decrease in protoplast volume, while the surface area of the vacuoles remained mostly unchanged. These morphometrical analyses suggest that the elliptical structures are reservoirs for excess VMs that result from the response to rapid decreases in vacuolar and protoplast volumes.

show abstract

“…The application of these schemes would however first require extracting the internal growth rings. Whereas the detection of the growth center and the detection of the external shape may be automated [6], the automated extraction of internal growth rings is a particularly complex task due to the presence of blind areas and so-called subjective contours [18,21]. As exemplified in Fig.…”

Section: Introduction and Problem Statementmentioning

confidence: 99%

2D Image-based reconstruction of shape deformation of biological structures using a level-set representation

Fablet¹,

Pujolle²,

Chessel³

et al. 2008

Computer Vision and Image Understanding

View full text Add to dashboard Cite

This paper copes with the reconstruction of accretionary growth sequence from images of biological structures depicting concentric ring patterns. Accretionary growth shapes are modeled as the level-sets of a potential function. Given an image of a biological structure, the reconstruction of the sequence of growth shapes is stated as a variational issue derived from geometric criteria. This variational setting exploits image-based information, in terms of the orientation field of relevant image structures, which leads to an original advection term. The resolution of this variational issue is discussed. Experiments on synthetic and real data are reported to validate the proposed approach.

show abstract

Autonomous agents for edge detection and continuity perception on otolith images

Cited by 16 publications

References 14 publications

Automated fish age estimation from otolith images using statistical learning

Automated fish age estimation from otolith images using statistical learning

Morphometrical study of plant vacuolar dynamics in single cells using three-dimensional reconstruction from optical sections

2D Image-based reconstruction of shape deformation of biological structures using a level-set representation

Contact Info

Product

Resources

About