A new and simple shape descriptor based on a non-parametric multiscale model (original) (raw)
Related papers
Global Contour and Region Based Shape Analysis and Similarity Measures
Defence Science Journal, 2013
More and more images have been generated in digital form around the world. There is a growing interest in finding images in large collections or from remote databases. In order to find an image, the image has to be described or represented by certain features. Shape is an important visual feature of an image. Searching for images using shape features has attracted much attention. There are many shape representation and description techniques in the literature. Object classification often operates by making decisions based on the values of several shape properties measured from an image of the object. Shape analysis is a useful tool for recognition of an object. This paper treats various aspects that are needed to solve shape matching problems: choosing the precise problem of global contour and region based shape analysis, selecting the properties of the similarity measure that are needed for the problem and choosing the specific similarity measure to compute the similarity.
Shape similarity measure based on correspondence of visual parts
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
ÐA cognitively motivated similarity measure is presented and its properties are analyzed with respect to retrieval of similar objects in image databases of silhouettes of 2D objects. To reduce influence of digitization noise, as well as segmentation errors, the shapes are simplified by a novel process of digital curve evolution. To compute our similarity measure, we first establish the best possible correspondence of visual parts (without explicitly computing the visual parts). Then, the similarity between corresponding parts is computed and aggregated. We applied our similarity measure to shape matching of object contours in various image databases and compared it to well-known approaches in the literature. The experimental results justify that our shape matching procedure gives an intuitive shape correspondence and is stable with respect to noise distortions. Index TermsÐShape representation, shape similarity measure, visual parts, discrete curve evolution. ae 1 INTRODUCTION A shape similarity measure useful for shape-based retrieval in image databases should be in accord with our visual perception. This basic property leads to the following requirements:
Contour segment-based shape descriptor
2011
A new descriptor model based on shape contour representation aimed to object recognition in non-structured images is presented. The model uses the second derivative of the contour curve –a rotation invariant representationfor contour representation. A contour segmentation criterion suitable for form description is presented. An adaptation of this descriptor able to operate in scale-space is discussed. Some results are presented and further research work is proposed involving descriptor’s use in more complex problems.
Lecture Notes in Computer Science, 2013
In this paper, an image shape retrieval method was evaluated using Euclidean, Intersect, Hamming and Cityblock distances and different kinds of k-nearest neighbours classifiers such as the original kNN, mean distance kNN and Weighted kNN. Shapes were described using a new method based on the description of the contour points, CPDH36R, obtaining better results than with the original CPDH shape descriptor. The efficiency in the retrieval was tested using Kimia99, Kimia25, MPEG7 and MPEG2 datasets obtaining an 84% of success rate in Kimia25, 94% in Kimia99, 91% in MPEG2 and 82% in MPEG7 datasets using our CPDH36R method, cityblock distance and original kNN against the 68%, 91%, 74% and 59% respectively obtained using the original CPDH. The greatest difference between the original method and our proposal can be seen clearly using MPEG2 dataset. Another advantage of our retrieval method, apart from the success rate, is the computational cost which is clearly better than the one achieved with the original Earth Mover Distance classifier used in the CPDH original method.
Robust shape matching using global feature space representation of contours
2012 International Conference on Computing, Networking and Communications (ICNC), 2012
The fundamental ingredient of content-based image retrieval is the selection of appropriate features to describe the content of the image. Shape of an object, represented by its contour, is the most important visual feature that is thought to be used by humans to determine the similarity of objects. In this paper, we present an effective representation of shape, using its boundary information, that is robust to arbitrary distortions and affine transformation. The contour representation of shape is converted into time series and is modeled using orthogonal basis function representations. Encoding contour representation of shapes in this manner leads to efficiency gains over existing approaches such as structural shape representation and techniques that use discrete point-based flow vectors to represent the contour. Experimental evaluation demonstrates that the proposed shape representation and matching mechanism is effective, efficient and robust to different arbitrary and affine distortions.
Coarse-to-fine multiscale affine invariant shape matching and classification
2004
In this paper, a multiscale algorithm for matching and classifying 2-D shapes is developed. The algorithm uses the 1-D Dyadic Wavelet Transform (DWT) to decompose a shape's boundary into multiscale levels. Then the coarse to fine matching and classification are achieved in two stages. In the first stage, the global features are extracted by calculating the curve moment invariants of the approximation coefficients. By calculating the normalized cross correlation of the 1-D triangle area representation of the detail coefficients, the local similarity is achieved by the second stage. The proposed algorithm is invariant to the affine transformation and to the boundary starting point variation. In addition, the results demonstrate that the new algorithm is not sensitive to small boundary deformations.
Nurbs: A new shape descriptor for shape-based image retrieval
2003
The representation, matching and analysis of objects of interest are of prime importance in shape-based retrieval systems. In the large image repositories, there arises a problem to find a set of images that are relevant to the user's needs. This necessitates an effective retrieval approach that resembles the human capability to retrieving images that are relevant to a query from a database. Perceiving a shape is to capture prominent elements of an object. For the purpose of retrieval by shape similarity, representation is preferred such that the salient perceptual aspects of a shape are captured and are able to imitate the human perception in perceiving shapes. The representation method is important, because the effectiveness of the representation will determine the accuracy of the retrieval results. Thus, there is a need to have an effective and accurate representation method. The representation features are then used to compute the similarity score between two images. In this paper, we present a new shape descriptor to represent all possible shapes using Non-Uniform Rational B-Spline (NURBS). We also present NURBS-Warping method, which is similar to elastic matching, to obtain similarity score in the retrieval process. We run two sets of experiments to show the efficiency of NURBS shape representation and NURBS-Warping method over B-Spline representation.