Content uploaded by Serge Beucher
Author content
All content in this area was uploaded by Serge Beucher on Sep 15, 2015
Content may be subject to copyright.
A preview of the PDF is not available
Segmentation: The Watershed Transformation. Mathematical Morphology in Image Processing
... After implementing the segmentation task, a specific portion of the image will be extracted for identifying a plant disease. Based on segmented regions, one can extract essential features such as color detail and boundaries for diagnosing diseases [7,8]. ...
Disease recognition in plants is one of the essential problems in agricultural image processing. This article focuses on designing a framework that can recognize and classify diseases on pomegranate plants exactly. The framework utilizes image processing techniques such as image acquisition, image resizing, image enhancement, image segmentation, ROI extraction (region of interest), and feature extraction. An image dataset related to pomegranate leaf disease is utilized to implement the framework, divided into a training set and a test set. In the implementation process, techniques such as image enhancement and image segmentation are primarily used for identifying ROI and features. An image classification will then be implemented by combining a supervised learning model with a support vector machine. The proposed framework is developed based on MATLAB with a graphical user interface. According to the experimental results, the proposed framework can achieve 98.39% accuracy for classifying diseased and healthy leaves. Moreover, the framework can achieve an accuracy of 98.07% for classifying diseases on pomegranate leaves.
... Subsequently, a postprocessing method 27,35 combined both images (edge and ROI) to produce the final binary segmentation image. Briefly, the edge image was smoothed based on the average cell size (automatically obtained by Fourier analysis) and subsequently the watershed algorithm 36 was applied to create the binary segmentation. Finally, cells at the image border or with less than 75% of their area within the ROI were discarded. ...
Purpose: To present a fully automatic method to estimate the corneal endothelium parameters from specular microscopy images and to use it to study a one-year follow-up after ultrathin Descemet stripping automated endothelial keratoplasty.
Methods: We analyzed 383 post ultrathin Descemet stripping automated endothelial keratoplasty images from 41 eyes acquired with a Topcon SP-1P specular microscope at 1, 3, 6, and 12 months after surgery. The estimated parameters were endothelial cell density (ECD), coefficient of variation (CV), and hexagonality (HEX). Manual segmentation was performed in all images.
Results: Our method provided an estimate for ECD, CV, and HEX in 98.4% of the images, whereas Topcon’s software had a success rate of 71.5% for ECD/CV and 30.5% for HEX. For the images with estimates, the percentage error in our method was 2.5% for ECD, 5.7% for CV, and 5.7% for HEX, whereas Topcon’s software provided an error of 7.5% for ECD, 17.5% for CV, and 18.3% for HEX. Our method was significantly better than Topcon’s (P < 0.0001) and was not statistically significantly different from the manual assessments (P > 0.05). At month 12, the subjects presented an average ECD = 1377 ± 483 [cells/mm2], CV = 26.1 ± 5.7 [%], and HEX = 58.1 ± 7.1 [%].
Conclusions: The proposed method obtains reliable and accurate estimations even in challenging specular images of pathologic corneas.
Translational relevance: CV and HEX, not currently used in the clinic owing to a lack of reliability in automatic methods, are useful biomarkers to analyze the postoperative healing process. Our accurate estimations allow now for their clinical use.
... One of the building blocks for the algorithms is the Watershed algorithm, as discussed by F. Meyer and S. Beucher in the work, 'The Morphological Approach to Segmentation: The Watershed Transformation' [7]. The work discusses in detail the intricacies of the algorithm, including the tools, transformations, uses, and the application of Watershed to images. ...
High-throughput phenotyping of seeds is the assessment of seed morphometry to aid in the prediction of yield, tolerance, resistance, and development of seeds in various environmental conditions. The paper focuses on the application of 3D graphics to image processing as a means to conduct seed phenotyping better. The paper proposes two algorithms-similar in the outcome, but different in implementation. The algorithms perform image processing on a variety of seeds such as wheat, soy, sorghum, rough rice, white rice, and canola to arrive at their morphometric estimations. In the area of static image processing, addressed are at least three common yet significant problems of seed clusters on images, skewed images, and poor image quality. As a means to address the problems, we propose the use of low-cost physical components. The algorithms provide the estimated count, area, perimeter, length, and width of seeds within an image.
... Other than that, a powerful tool method called Watershed Transform, which is based on the region segmentation method has been widely used in medical images. The original watershed method was first done by Lantuejoul in 1978 [20], and its applications have been widely described by Beucher and Meyer [21]. In this paper, a gradient image of panoramic dental radiograph is computed based on the convolution of the original image with a Sobel filter. ...
... The distance map method works well for spherical grains since the centre of a sphere is a local maximum of the distance from the edge of the grains. However, for the non-spherical lactose grains, the distance map could be seen through visual inspection to consistently over-segment particles (Beucher and Meyer, 1993), i.e. ...
Carrier-based dry powder inhaler (DPI) formulations need to be accurately characterised for their particle size distributions, surface roughnesses, fines contents and flow properties. Understanding the micro-structure of the powder formulation is crucial, yet current characterisation methods give incomplete information. Commonly used techniques like laser diffraction (LD) and optical microscopy (OM) are limited due to the assumption of sphericity and can give variable results depending on particle orientation and dispersion. The aim of this work was to develop new powder analytical techniques using X-ray computed tomography (XCT) that could be employed for non-destructive metrology of inhaled formulations. α-lactose monohydrate powders with different characteristics have been analysed, and their size and shape (sphericity/aspect ratio) distributions compared with results from LD and OM. The three techniques were shown to produce comparable size distributions, while the different shape distributions from XCT and OM highlight the difference between 2D and 3D imaging. The effect of micro-structure on flowability was also analysed through 3D measurements of void volume and tap density. This study has demonstrated for the first time that XCT provides an invaluable, non-destructive and analytical approach to obtain number- and volume-based particle size distributions of DPI formulations in 3D space, and for unique 3D characterisation of powder micro-structure.
... The algorithm circumscribed regions of interests (ROI) in the vicinity of the probe track in four (anterior, posterior, medial, lateral) directions, in accordance with previous studies [34,35] (Fig. 2D). The cell centroids in NeuN-immunostained section images were detected with an algorithm based on the watershed method [40], which allows the discrimination of cell bodies located close to each other. The number of centroids was calculated in 20 μm wide sectors up to a distance of 400 μm from the track, as visible on the generated Matlab figure ( Fig. 2A-D). ...
The use of SU-8 material in the production of neural sensors has grown recently. Despite its widespread application, a detailed systematic quantitative analysis concerning its biocompatibility in the central nervous system is lacking. In this immunohistochemical study, we quantified the neuronal preservation and the severity of astrogliosis around SU-8 devices implanted in the neocortex of rats, after a 2 months survival.
We found that the density of neurons significantly decreased up to a distance of 20 μm from the implant, with an averaged density decrease to 24 ± 28% of the control. At 20 to 40 μm distance from the implant, the majority of the neurons was preserved (74 ± 39% of the control) and starting from 40 μm distance from the implant, the neuron density was control-like. The density of synaptic contacts – examined at the electron microscopic level – decreased in the close vicinity of the implant, but it recovered to the control level as close as 24 μm from the implant track. The intensity of the astroglial staining significantly increased compared to the control region, up to 560 μm and 480 μm distance from the track in the superficial and deep layers of the neocortex, respectively. Electron microscopic examination revealed that the thickness of the glial scar was around 5–10 μm thin, and the ratio of glial processes in the neuropil was not >16% up to a distance of 12 μm from the implant.
Our data suggest that neuronal survival is affected only in a very small area around the implant. The glial scar surrounding the implant is thin, and the presence of glial elements is low in the neuropil, although the signs of astrogliosis could be observed up to about 500 μm from the track. Subsequently, the biocompatibility of the SU-8 material is high. Due to its low cost fabrication and more flexible nature, SU-8 based devices may offer a promising approach to experimental and clinical applications in the future.
Directed acyclic graphs (DAGs) are used in many domains ranging from computer science to bioinformatics, including industry and geoscience. They enable to model complex evolutions where spatial objects (e.g., soil erosion) may move, (dis)appear, merge or split. We study a new graph-based representation, called attributed DAG (a-DAG). It enables to capture interactions between objects as well as information on objects (e.g., characteristics or events). In this paper, we focus on pattern mining in such data. Our patterns, called weighted paths, offer a good trade-off between expressiveness and complexity. Frequency and compactness constraints are used to filter out uninteresting patterns. These constraints lead to an exact condensed representation (without loss of information) in the single-graph setting. A depth-first search strategy and an optimized data structure are proposed to achieve the efficiency of weighted path discovery. It does a progressive extension of patterns based on database projections. Relevance, scalability and genericity are illustrated by means of qualitative and quantitative results when mining various real and synthetic datasets. In particular, we show how such an approach can be used to monitor soil erosion using remote sensing and geographical information system (GIS) data.
This article models multicolor perceptual features to investigate human perceptual responses to multiple colors based on a subjective evaluation experiment. First, an image set containing 164 multicolor samples was constructed, and multicolor objective features containing two categories with 60 dimensions were designed and quantified. Based on this, the 164 samples were evaluated by 30 subjects based on nine perceptual descriptive variables, including “swell‐shrink,” “light‐dark,” and “far‐near.” Then, a mathematical model of multicolor perceptual features (nine descriptive variables) was established by support vector regression. Finally, this article conducted model evaluation and factor analysis, and the results showed that the model could predict multicolor perceptual features well. It was also observed that, unlike single‐color perception, space perception was produced when applying multiple colors.
Methods for image segmentation using mathematical morphology are presented. These methods are based on two main tools: the watershed transform and the homotopy modification which solve the problem of the oversegmentation and introduce the notion of markers of the objects to be segmented in the image. Some examples in various domains (biology, medicine, scene analysis, 3D images, detection of moving objects, color images) are given. We tried in these examples to emphasize the problems encountered and to explain shortly the proposed solutions. The algorithms are given in the Appendix. The detection of the markers requiring a large amount of morphological tools, many of them, are presented, though not directly related to segmentation
For an optical or acoustical wavefront running through a medium of space variant refraction index the eikonal equation connects local front arrival time with local refraction index. So-called difference approximation methods are known for solving the spatial wavefront development with time and thus, indirectly, the eikonal equation.Here a novel fast method for the calculation of an approximative solution of the eikonal equation is proposed.From literature it is known that by solving an eikonal equation one can construct a line pattern rendition of a given image. We have generalized this method and made it fit for line engravings.We have found yet another kind of image display based on solving an eikonal equation: shading from shape. We propose to construct a matte 3-D surface (shape) that, when illuminated perpendicularly and imaged in eye or camera, yields a grey value (shading, luminance) field that renders the image.Both methods have been applied in a recent design for a Dutch coin.
Based on a two-component descriptor, a distance label for each point, it is shown that Euclidean distance maps can be generated by effective sequential algorithms. The map indicates, for each pixel in the objects (or the background) of the originally binary picture, the shortest distance to the nearest pixel in the background (or the objects). A map with negligible errors can be produced in two picture scans which has to include forward and backward movement for each line. Thus, for expanding/shrinking purposes it may compete very successfully with iterative parallel propagation in the binary picture itself. It is shown that skeletons can be produced by simple procedures and since these are based on Euclidean distances it is assumed that they are superior to skeletons based on d4−, d8−, and even octagonal metrics.
Based on the notion of homotopy, a complete theory of the skeleton in digital spaces is given. We derive from it several very fast and efficient algorithms for constructing the skeleton (sequential or contour following algorithms. In this way we obtain easily, for all types of digital grids, all transformations which belong to the same family as the skeleton: binary or grey tone skeletons, euclidian or geodesic skeletons, minimal skeletons or perceptual graphs, conditional bisectors, etc. The properties of the resulting skeleton make it a good candidate for being the correct discretization of the skeletons in the continuous plane.