P. Nagabhushan

A simple and robust line detection algorithm based on small eigenvalue analysis

In this paper, a simple and robust algorithm is proposed for detecting straight line segments in an edge image. The proposed algorithm is based on small eigenvalue analysis. The statistical and geometrical properties of the small eigenvalue of the covariance matrix of a set of edge pixels over a connected region of support are explored for the purpose of straight line detection. The approach scans an input edge image from the top left corner to the bottom right corner with a moving mask of size k × k for some odd integer k > 1. At every stage, the small eigenvalue of the covariance matrix of the edge pixels covered by the mask and connected to the center pixel of the mask is computed. These pixels are said to be linear edge pixels if the computed small eigenvalue is less than a pre-defined threshold value. Several experiments have been conducted on various images with considerable background noise and also with significant edge point location errors to reveal the efficacy of the proposed model. The results of the experiments emphasize that the proposed model outperforms other models specifically the Hough transform and its variants in addition to being robust to image transformations such as rotation, scaling and translation.

Multivalued type proximity measure and concept of mutual similarity value useful for clustering symbolic patterns

In this paper, a novel similarity measure for estimating the degree of similarity between two patterns (described by interval type data) is proposed. The proposed measure computes the degree of similarity between two patterns and approximates the computed similarity value by a multivalued type data. Unlike conventional proximity matrices, the proximity matrix obtained through the application of the proposed similarity measure is not necessarily symmetric. Based on this unconventional similarity matrix a modified agglomerative method by introducing the concept of mutual similarity value (MSV) for clustering symbolic patterns is also presented. Experiments on various data sets have been conducted in order to study the efficacy of the proposed methodology.

Letters: (2D)2 FLD: An efficient approach for appearance based object recognition

In this paper, a new technique called 2-directional 2-dimensional Fishers Linear Discriminant analysis ((2D)^2 FLD) is proposed for object/face image representation and recognition. We first argue that the standard 2D-FLD method works in the row direction of images and subsequently we propose an alternate 2D-FLD which works in the column direction of images. To straighten out the problem of massive memory requirements of the 2D-FLD method and as well the alternate 2D-FLD method, we introduce (2D)^2 FLD method. The introduced (2D)^2 FLD method has the advantage of higher recognition rate, lesser memory requirements and better computing performance than the standard PCA/2D-PCA/2D-FLD method, and the same has been revealed through extensive experimentations conducted on COIL-20 dataset and AT&T face dataset.

Adaptive dissimilarity index for measuring time series proximity

The most widely used measures of time series proximity are the Euclidean distance and dynamic time warping. The latter can be derived from the distance introduced by Maurice Fréchet in 1906 to account for the proximity between curves. The major limitation of these proximity measures is that they are based on the closeness of the values regardless of the similarity w.r.t. the growth behavior of the time series. To alleviate this drawback we propose a new dissimilarity index, based on an automatic adaptive tuning function, to include both proximity measures w.r.t. values and w.r.t. behavior. A comparative numerical analysis between the proposed index and the classical distance measures is performed on the basis of two datasets: a synthetic dataset and a dataset from a public health study.

A new scheme for unconstrained handwritten text-line segmentation

Variations in inter-line gaps and skewed or curled text-lines are some of the challenging issues in segmentation of handwritten text-lines. Moreover, overlapping and touching text-lines that frequently appear in unconstrained handwritten text documents significantly increase segmentation complexities. In this paper, we propose a novel approach for unconstrained handwritten text-line segmentation. A new painting technique is employed to smear the foreground portion of the document image. The painting technique enhances the separability between the foreground and background portions enabling easy detection of text-lines. A dilation operation is employed on the foreground portion of the painted image to obtain a single component for each text-line. Thinning of the background portion of the dilated image and subsequently some trimming operations are performed to obtain a number of separating lines, called candidate line separators. By using the starting and ending points of the candidate line separators and analyzing the distances among them, related candidate line separators are connected to obtain segmented text-lines. Furthermore, the problems of overlapping and touching components are addressed using some novel techniques. We tested the proposed scheme on text-pages of English, French, German, Greek, Persian, Oriya and Bangla and remarkable results were obtained.

Triangular spatial relationship: a new approach for spatial knowledge representation

Abstract In this paper, an important aspect of problems in symbolic image database retrieval (SIDR) is addressed. The problems in conventional pairwise spatial relationships, particularly the 9DLT-matrix-based approach, used to represent a symbolic image in SID are discussed. A new concept called triangular spatial relationship (TSR) is introduced as an improved, invariant spatial relationship to take care of object transformations. A set of quadruples is used to preserve the TSR among the components in a symbolic image. The proposed method is based on principal component analysis. The first principal component vector (PCV) of the set of quadruples is used to represent a symbolic image in SID. The proposed algorithm requires O(log n ) search time in the worst case in the recognition phase, where n is the number of symbolic images stored in the symbolic image database.

Modified region decomposition method and optimal depth decision tree in the recognition of non-uniform sized characters - An experimentation with Kannada characters

Abstract In contrast to English alphabets, some characters in Indian languages such as Kannada, Hindi, Telugu may have either horizontal or vertical or both the extensions making it difficult to enclose every such character in a standard rectangular grid as done quite often in character recognition research. In this work, an improved method is proposed for the recognition of such characters (especially Kannada characters), which can have spread in vertical and horizontal directions. The method uses a standard sized rectangle which can circumscribe standard sized characters. This rectangle can be interpreted as a two-dimensional, 3×3 structure of nine parts which we define as bricks. This structure is also interpreted as consecutively placed three row structures of three bricks each or adjacently placed three column structures of three bricks each. It is obvious that non-uniform sized characters cannot be contained within the standard rectangle of nine bricks. The work presented here proposes to take up such cases. If the character has horizontal extension, then the rectangle is extended horizontally by adding one column structure of three bricks at a time, until the character is encapsulated. Likewise, for vertically extended characters, one row structure is added at a time. For the characters which are smaller than the standard rectangle, one column structure is removed at a time till the character fits in the shrunk rectangle. Thus, the character is enclosed in a rectangular structure of m×n bricks where m⩾3 and n⩾1 . The recognition is carried out intelligently by examining certain selected bricks only instead of all mn bricks. The recognition is done based on an optimal depth logical decision tree developed during the Learning phase and does not require any mathematical computation.

Dimensionality reduction of symbolic data

Abstract Hitherto dimensionality/feature reduction techniques are studied with reference to conventional data, where the objects are represented by numerical vectors. This proposal is to extend the notion of dimensionality reduction to more generalised objects called Symbolic data. A mathematical model which achieves generation of symbolic features — particularly of span type — in transformed lower-dimensional space from a high n -dimensional feature space of span type symbolic data, is presented in this paper. This work is expected to open a new avenue in the area of symbolic data analysis.

A three-dimensional neural network model for unconstrained handwritten numeral recognition : A new approach

The paper describes a three-dimensional (3-D) neural network recognition system for conflict resolution in recognition of unconstrained handwritten numerals. This neural network classifier is a combination of modified self-organizing map (MSOM) and learning vector quantization (LVQ). The 3-D neural network recognition system has many layers of such neural network classifiers and the number of layers forms the third dimension. The Experiments are conducted employing SOM, MSOM, SOM and LVQ, and MSOM and LVQ networks. These experiments on a database of unconstrained handwritten samples show that the combination of MSOM and LVQ performs better than other networks in terms of classification, recognition and training time. The 3-D neural network eliminates the substitution error.

Archival and retrieval of symbolic images: an invariant scheme based on triangular spatial relationship

In this paper, a novel scheme for representing symbolic images in a symbolic image database (SID) is proposed. The proposed scheme is based on triangular spatial relationship (TSR) [Pattern Recognition Lett. 22 (2001) 999]. The scheme preserves TSR among the components in a symbolic image by the use of quadruples. A SID is created through the construction of B-tree, an efficient multilevel indexing structure. A methodology to retrieve similar images for a given query image is also presented. The presented retrieval model has logarithmic search time complexity. The study made in this work reveals that the model bears various advantages when compared to other existing models and could be extended towards dynamic databases.