Didier Coquin

Dynamic hand gesture recognition using the skeleton of the hand

This paper discusses the use of the computer vision in the interpretation of human gestures. Hand gestures would be an intuitive and ideal way of exchanging information with other people in a virtual space, guiding some robots to perform certain tasks in a hostile environment, or interacting with computers. Hand gestures can be divided into two main categories: static gestures and dynamic gestures. In this paper, a novel dynamic hand gesture recognition technique is proposed. It is based on the 2D skeleton representation of the hand. For each gesture, the hand skeletons of each posture are superposed providing a single image which is the dynamic signature of the gesture. The recognition is performed by comparing this signature with the ones from a gesture alphabet, using Baddeleys distance as a measure of dissimilarities between model parameters.

A parametric active polygon for leaf segmentation and shape estimation

In this paper we present a system for tree leaf segmentation in natural images that combines a first, unrefined segmentation step, with an estimation of descriptors depicting the general shape of a simple leaf. It is based on a light polygonal model, built to represent most of the leaf shapes, that will be deformed to fit the leaf in the image. Avoiding some classic obstacles of active contour models, this approach gives promising results, even on complex natural photographs, and constitutes a solid basis for a leaf recognition process.

Understanding leaves in natural images - A model-based approach for tree species identification

With the aim of elaborating a mobile application, accessible to anyone and with educational purposes, we present a method for tree species identification that relies on dedicated algorithms and explicit botany-inspired descriptors. Focusing on the analysis of leaves, we developed a working process to help recognize species, starting from a picture of a leaf in a complex natural background. A two-step active contour segmentation algorithm based on a polygonal leaf model processes the image to retrieve the contour of the leaf. Features we use afterwards are high-level geometrical descriptors that make a semantic interpretation possible, and prove to achieve better performance than more generic and statistical shape descriptors alone. We present the results, both in terms of segmentation and classification, considering a database of 50 European broad-leaved tree species, and an implementation of the system is available in the iPhone application Folia.

Discrete distance operator on rectangular grids

Abstract In this paper we present a new local distance transformation adapted to rectangular grids. This situation occurs with most industrial vision systems. Such operators allow Euclidean distance transform images to be approximated by using only local operations. These operators are optimized in the context of minimizing the maximum error over circular trajectories. The formulas of the coefficients as a function of the pixel width are given in the case of a 5 × 5 neighborhood. Experimental results and comparisons with other distance operators are then presented.

Application of Baddeley's distance to dissimilarity measurement between gray scale images

Abstract In this paper, we introduce a dissimilarity measure between two gray-scale images based on Baddeleys distance. To some extent, it can be regarded as a modification of that proposed by Wilson et al. [Internat. J. Comput. Vision 24 (1) (1997) 5–18]. Images are represented by surfaces in a 3D space, instead of their subgraphs. Distance calculations are performed by means of a 3D local distance operator adapted to parallelepipedic grids. No truncation effect is introduced. Properties of the new dissimilarity operator are compared to those of the Wilson–Baddeley–Owen operator in terms of sensitivity to gray level variations, spatial shifts and shape distortions. Compared with the Wilson–Baddeley–Owen operator, a more linear behavior is observed. A simplified phenomenological model is proposed in order to explain this behavior.

Fuzzy color-based approach for understanding animated movies content in the indexing task

This paper proposes a method for detecting and analyzing the color techniques used in the animated movies. Each animated movie uses a specific color palette which makes its color distribution one major feature in analyzing the movie content. The color palette is specially tuned by the author in order to convey certain feelings or to express artistic concepts. Deriving semantic or symbolic information from the color concepts or the visual impression induced by the movie should be an ideal way of accessing its content in a content-based retrieval system. The proposed approach is carried out in two steps. The first processing step is the low-level analysis. The movie color content gets represented with several global statistical parameters computed from the movie global weighted color histogram. The second step is the symbolic representation of the movie content. The numerical parameters obtained from the first step are converted into meaningful linguistic concepts through a fuzzy system. They concern mainly the predominant hues of the movie, some of Ittens color contrasts and harmony schemes, color relationships and color richness. We use the proposed linguistic concepts to link to given animated movies according to their color techniques. In order to make the retrieval task easier, we also propose to represent color properties in a graphical manner which is similar to the color gamut representation. Several tests have been conducted on an animated movie database.

A skeletonization algorithm using chamfer distance transformation adapted to rectangular grids

Many skeletonization methods based on distance transformation have been proposed. They are generally designed for square grids. However most industrial vision systems digitize images on rectangular grids. These images are then resampled to obtain square pixels. In this paper we propose a skeletonization algorithm directly adapted to rectangular grids. We use the efficient skeletonization method of Arcelli and Sanniti di Baja (1993), based on the Euclidean distance transform, and we adapt it to chamfer distances computed on rectangular grids. We propose some modifications to this algorithm. Some results are given and discussed.

A model-based approach for compound leaves understanding and identification

In this paper, we propose a specific method for the identification of compound-leaved tree species, with the aim of integrating it in an educational smartphone application. Our work is based on dedicated shape models for compound leaves, designed to estimate the number and shape of leaflets. A deformable template approach is used to fit these models and produce a high-level interpretation of the image content. The resulting models are later used for the segmentation of leaves in both plain and natural background images, by the use of multiple region-based active contours. Combined with other botany-inspired descriptors accounting for the morphological properties of the leaves, we propose a classification method that makes a semantic interpretation possible. Results are presented over a set of more than 1000 images from 17 European tree species, and an integration in the existing mobile application Folia1 is considered.

Leaf margins as sequences

We propose a high-level interpretation of leaf contours through CSS analysis.Leaf margins are described by an original string representation of vectorial symbols.Methods inspired from string processing are adapted to this objects.We use this spatially-rich descriptor to perform species classification.Performance is compared with state-of-the-art shape descriptors, and shows competitive results. In the context of an automated leaf identification process, the use of thorough leaf margin descriptors is essential given the importance of this criterion in the determination of the species. The spatial properties of teeth along the leaf contour are something to keep track of, which is made possible through the use of structured representations. This paper introduces a sequence representation of leaf margins where teeth are viewed as symbols of a multivariate real valued alphabet. It presents the methods developed to make use of this description for classification and implementation in a mobile tree identifying application. The results of various classification methods are compared and discussed, both in terms of species recognition and of consistency with botanical concepts.

Fuzzy semantic action and color characterization of animation movies in the video indexing task context

This paper presents a fuzzy statistical approach for the semantic content characterization of the animation movies. The movie action content and color properties play an important role in the understanding of the movie content, being related to the artistic signature of the author. That is why the proposed approach is carried out by analyzing several statistical parameters which are computed both from the movie shot distribution and the global color distribution. The first category of parameters represents the movie mean shot change speed, the transition ratio and the action ratio while the second category represents the color properties in terms of color intensity, warmth, saturation and color relationships. The semantic content characterizations are achieved from the low-level parameters using a fuzzy representation approach. Hence, the movie content is described in terms of action, mystery, explosivity, predominant hues, color contrasts and the color harmony schemes. Several experimental tests were performed on an animation movie database. Moreover, a classification test was conducted to prove the discriminating power of the proposed semantic descriptions for their prospective use as semantic indexes in a content-based video retrieval system.