Marta Zamfir

Color morphology-like operators based on color geometric shape characteristics

Mathematical morphology is based on two infimum- and respectively, supremum-commuting operations (the erosion and the dilation). In the scalar case, these operations are obviously the minimum and maximum. In the vector-valued case, minimum and maximum cannot be easily defined. Pixels within color images are described by three-component vectors, and thus the mathematical morphology is difficult to introduce for colors. We propose pseudo-morphology based on reduced ordering of colors (associate a scalar to each color, order the scalars and impose their ranking to their corresponding colors). The approach has been widely investigated, by proposing different scalars (usually the same scalars as used for distance-based color image filtering). We propose the use of scalars issued as geometrical shape invariants for a triangle-representation of colors.

A Study of the Influence of the PSF Accuracy on the Quality of Image Deblurring

The paper presents a few recent results obtained in restoration of motion blurred photographs. Qualitative differences between images restored with point spread functions of different shapes, values or sizes are illustrated, using test images and simulated blur.

Automatic segmentation of infantile hemangiomas

Infantile hemangiomas are the most common types of tumors that are found in infants and have an incidence of approximately 10% in the common population. Although most infantile hemangiomas are self-involuting, due to their fast proliferation they may threaten vital anatomical structures and physiological functions; also, the involution process may take up to several years. An accurate monitoring of the progress of hemangioma growth and regression is essential. We thus suggest using a computer aided follow-up monitoring of these lesions by an automatic detection and quantifying the lesion dynamic: regression or proliferation. In this study we propose some image enhancements methods and also a preliminary color based segmentation. We have tested our methods on 25 hemangioma cases and compared the automated segmentation results with clinician-determined segmentation using an area percentage error.

Nonlinear color image filtering by color to planar shape mapping

Nonlinear rank-order-based filtering of color images is difficult to implement; the multivariate nature of colors does not allow the introduction of a mathematically-correct and topology-preserving ordering relation. The most widely investigated approach is based on the use of reduced ordering relations introduced according to different scalars. We propose pseudo-morphologic and median operators based on reduced ordering of colors, with respect to scalars computed as geometrical shape invariants of a triangle representation of colors. The same color representation allows the introduction of a luminance invariant intercolor distance, used with good results in distance-based color filters.

Automatic monitoring system for the detection and evaluation of the evolution of hemangiomas

In this paper we introduce an automatic monitoring system for the detection and the evaluation of the evolution of hemangiomas using a fuzzy logic system based on two parameters: area and redness. We have considered pairs of images (from two different moments in time) that show hemangiomas either evolving, stationary or regressing. The starting points of the algorithm are the rectangular regions of interest (ROI), manually selected for each of the two images, and automatically segmented using Fuzzy C-means. Using the area and the redness of the hemagiomas extracted with Fuzzy C-means, for the same patient, at different moments of time, the algorithm decides whether the hemangioma is evolving, stationary or regressing. Because it is also useful to understand how the tumors shape is changing in time, we have also included a method of matching and overlapping the hemangioma regions.

Correlating human color similarity judgments and colorimetric representations

A color similarity test was conducted on the 24 color patches of a Gretag Macbeth color checker. Color similarities were measured either by distances between standard colorimetric representations (such as RGB, Lab or spectral reflectance curves) or by human observer judgments. In each case, the dissimilarity matrix was processed by a classical, metric, multidimensional scaling algorithm, in order to produce a visually-interpretable two-dimensional plot of color dissimilarity. The analysis of the plots produces some interesting conclusions. First, the plots produced by the Lab, RGB and spectral representations exhibit very evident variation axes according to the luminance and basic chromatic differences (red-green, blue-yellow). This behavior (trivial for the Lab representation) suggests that the color similarity measurement by chromatic differences is implicitly embedded in the RGB and spectral representations. The color dissimilarity plots associated to the human judgments (for any individual, as well as for an “average” observer) exhibit a different organization, which mixes hue, saturation and luminance (HSV). According to these plots, the human similarity judgment is not entirely HSV-based. We prove that it is possible to obtain the same color dissimilarity plots if a fuzzy color model is assumed. The fuzzy color model provides similarity coefficients (similarity degrees) between pairs of colors, based on their inter-distance, according to an imposed “color confusion” control parameter, which seems to be relevant for the human vision.

Automatic Detection of Hemangioma through a Cascade of Self-organizing Map Clustering and Morphological Operators

Abstract In this paper we propose a method for the automatic detection of hemangioma regions, consisting of a cascade of algorithms: a Self Organizing Map (SOM) for clustering the image pixels in 25 classes (using a 5x5 output layer) followed by a morphological method of reducing the number of classes (MMRNC) to only two classes: hemangioma and non-hemangioma. We named this method SOM-MMRNC. To evaluate the performance of the proposed method we have used Fuzzy C-means (FCM) for comparison. The algorithms were tested on 33 images; for most images, the proposed method and FCM obtain similar overall scores, within one percent of each other. However, in about 18% of the cases, there is a significant increase of the overall score for SOM-MMRNC (over 3.5%). On average, the results obtained with the proposed cascade are 1.06% better for each image.

Non-linear color image processing by color to plane polygon association

Non-linear rank-order-based filtering of color images is difficult to implement; the multivariate nature of colors does not allow the introduction of a mathematically-correct and topology-preserving ordering relation. The most widely investigated approach is based on the use of reduced ordering relations introduced according to different scalars. We propose pseudo-morphologic operators based on the reduced ordering of colors, with respect to scalars computed as geometrical shape invariants of a plane, polygonal representation of colors.

A new inter-color distance based on color to planar shape mapping

Although simple and based on physical realizable color primaries, the RGB color space cannot allow the direct definition of a topology-preserving, perceptually-compatible inter-color distance. The classical normalized inter-color distances are rather complex to implement and tune. We propose the use of a planar color representation plot, derived from multivariate data visualization, which allows the introduction of a luminance-invariant, geometry-based inter-color distance, used with good results in the implementation of distance-based color filters.