Automatic text block separation in document images

Separation of printed text blocks from the non-text areas, containing signatures, handwritten text, logos and other such symbols, is a necessary first step for an OCR involving printed text recognition. In the present work, we compare the efficacy of some feature-classifier combinations to carry out this separation task. We have selected length-nomalized horizontal projection profile (HPP) as the starting point of such a separation task. This is with the assumption that the printed text blocks contain lines of text which generate HPP's with some regularity. Such an assumption is demonstrated to be valid. Our features are the HPP and its two transformed versions, namely, eigen and Fisher profiles. Four well known classifiers, namely, Nearest neighbor, Linear discriminant function, SVM's and artificial neural networks have been considered and efficiency of the combination of these classifiers with the above features is compared. A sequential floating feature selection technique has been adopted to enhance the efficiency of this separation task. The results give an average accuracy of about 96.

Protein Block Expert (PBE): a web-based protein structure analysis server using a structural alphabet

Encoding protein 3D structures into 1D string using short structural prototypes or structural alphabets opens a new front for structure comparison and analysis. Using the well-documented 16 motifs of Protein Blocks (PBs) as structural alphabet, we have developed a methodology to compare protein structures that are encoded as sequences of PBs by aligning them using dynamic programming which uses a substitution matrix for PBs. This methodology is implemented in the applications available in Protein Block Expert (PBE) server. PBE addresses common issues in the field of protein structure analysis such as comparison of proteins structures and identification of protein structures in structural databanks that resemble a given structure. PBE-T provides facility to transform any PDB file into sequences of PBs. PBE-ALIGNc performs comparison of two protein structures based on the alignment of their corresponding PB sequences. PBE-ALIGNm is a facility for mining SCOP database for similar structures based on the alignment of PBs. Besides, PBE provides an interface to a database (PBE-SAdb) of preprocessed PB sequences from SCOP culled at 95% and of all-against-all pairwise PB alignments at family and superfamily levels. PBE server is freely available at http://bioinformatics.univ-reunion.fr/ PBE/.

Hexagonally ordered arrays of metallic nanodots from thin films of functional block copolymers

We demonstrate a new and simple route to fabricate highly dense arrays of hexagonally close packed inorganic nanodots using functional diblock copolymer (PS-b-P4VP) thin films. The deposition of pre-synthesized inorganic nanoparticles selectively into the P4VP domains of PS-b-P4VP thin films, followed by removal of the polymer, led to highly ordered metallic patterns identical to the order of the starting thin film. Examples of Au, Pt and Pd nanodot arrays are presented. The affinity of the different metal nanoparticles towards P4VP chains is also understood by extending this approach to PS-b-P4VP micellar thin films. The procedure used here is simple, eco-friendly, and compatible with the existing silicon-based technology. Also the method could be applied to various other block copolymer morphologies for generating 1-dimensional (1D) and 2-dimensional (2D) structures. (c) 2010 Elsevier Ltd. All rights reserved.

Non-periodic tilings in 2-dimensions with 4, 6, 8, 10 and 12-fold symmetries

The two dimensional plane can be filled with rhombuses, so as to generate non-periodic tilings with 4, 6, 8, 10 and 12-fold symmetries. Some representative tilings constructed using the rule of inflation are shown. The numerically computed diffraction patterns for the corresponding tilings are also shown to facilitate a comparison with possible X-ray or electron diffraction pictures.

STBCs with Reduced Sphere Decoding Complexity for Two-User MIMO-MAC

In this paper, Space-Time Block Codes (STBCs) with reduced Sphere Decoding Complexity (SDC) are constructed for two-user Multiple-Input Multiple-Output (MIMO) fading multiple access channels. In this set-up, both the users employ identical STBCs and the destination performs sphere decoding for the symbols of the two users. First, we identify the positions of the zeros in the R matrix arising out of the Q-R decomposition of the lattice generator such that (i) the worst case SDC (WSDC) and (ii) the average SDC (ASDC) are reduced. Then, a set of necessary and sufficient conditions on the lattice generator is provided such that the R matrix has zeros at the identified positions. Subsequently, explicit constructions of STBCs which results in the reduced ASDC are presented. The rate (in complex symbols per channel use) of the proposed designs is at most 2/N-t where N-t denotes the number of transmit antennas for each user. We also show that the class of STBCs from complex orthogonal designs (other than the Alamouti design) reduce the WSDC but not the ASDC.

X-ray diffraction from single Al6CuLi3 grains showing five-fold symmetry

We present here the detailed results of X-ray diffraction from single quasicrystals of Al6CuLi3. X-ray precession photographs taken down the two-, three- and five-fold axes along with rotation and zero-level Weissenberg photographs are shown. Preliminary analysis of the diffraction data rules out the twin hypothesis.

Non-periodic tilings in 2-dimensions: 4- and 7-fold symmetries

By inflating basic rhombuses, with a self-similarity principle, non-periodic tiling of 2-d planes is possible with 4, 5, 6, 7, 8, … -fold symmetries. As examples, non-periodic tilings with crystallographically allowed 4-fold symmetry and crystallographically forbidden 7-fold symmetry are presented in detail. The computed diffraction patterns of these tilings are also discussed.

Novel rhodanine derivatives induce growth inhibition followed by apoptosis

We have designed and synthesized three novel compounds, 5-isopropylidiene derivatives of 3-dimethyl-2-thio-hydantoin (ITH-1), 3-ethyl-2-thio-2,4-oxazolidinedione (ITO-1), and 5-benzilidene-3-ethyl rhodanine (BTR-1), and have tested their chemotherapeutic properties. Our results showed that all three compounds induced cytotoxicity in a time-and concentration-dependent manner on leukemic cell line, CEM. Among the compounds tested, BTR-1 was 5- to 7-fold more potent than ITH-1 and ITO-1 when compared by trypan blue and MTT assays. IC50 value of BTR-1 was estimated to be <10 mu M. Both cell cycle analysis and tritiated thymidine assays revealed that BTR-1 affects DNA replication by inducing a block at S phase. BTR-1 treatment led to increased level of ROS production and DNA strand breaks suggesting activation of apoptosis for induction of cell death. (C) 2010 Elsevier Ltd. All rights reserved.

A Vicarious One-Pot Transformation of a Fluorocyclitol to a Polycyclitol: Observation of a Formal 4-Fold Axial-Equatorial Epimerization on a Conformationally Locked Scaffold

A conformationally locked fluoropentol undergoes an interesting transformation to (trans,anti,trans,anti,trans)-perhydro-2,3,4a,6,7,8a-naphthalenehexol essentially under conditions of base-induced transesterification. The proposed rationale for the observed metamorphosis involves a nucleophilic displacement of fluoride, and subsequent stereo- and regioselective anti-Furst-Plattner-type ring-opening of the epoxide thus formed.

Five-fold quasiperiodic tilings generated by inflated rhombuses with identical boundaries: possible application to the growth of quasicrystals

In order to generate normal Penrose tilings by inflation/deflation, decisions have to be made regarding the matching of the rhombuses/tilings with their neighbours. We show here that this decision-making problem can be avoided by adopting a deflation/inflation procedure which uses the decorated rhombuses with identical boundaries. The procedure enables both kinds of inflated rhombuses to match in any orientation along their edges. The tilings so generated are quasiperiodic. These structures appear to have a close relationship with the growth mechanism of quasicrystals.

On Computation of Minimum Distance of Linear Block Codes Above 1/2 Rate Coding

The minimum distance of linear block codes is one of the important parameter that indicates the error performance of the code. When the code rate is less than 1/2, efficient algorithms are available for finding minimum distance using the concept of information sets. When the code rate is greater than 1/2, only one information set is available and efficiency suffers. In this paper, we investigate and propose a novel algorithm to find the minimum distance of linear block codes with the code rate greater than 1/2. We propose to reverse the roles of information set and parity set to get virtually another information set to improve the efficiency. This method is 67.7 times faster than the minimum distance algorithm implemented in MAGMA Computational Algebra System for a (80, 45) linear block code.

Reduced ML-Decoding Complexity, Full-Rate STBCs for 4 Transmit Antenna Systems

For an n(t) transmit, n(r) receive antenna system (n(t) x nr system), a full-rate space time block code (STBC) transmits min(n(t), n(r)) complex symbols per channel use. In this paper, a scheme to obtain a full-rate STBC for 4 transmit antennas and any n(r), with reduced ML-decoding complexity is presented. The weight matrices of the proposed STBC are obtained from the unitary matrix representations of a Clifford Algebra. By puncturing the symbols of the STBC, full rate designs can be obtained for n(r) < 4. For any value of n(r), the proposed design offers the least ML-decoding complexity among known codes. The proposed design is comparable in error performance to the well known Perfect code for 4 transmit antennas while offering lower ML-decoding complexity. Further, when n(r) < 4, the proposed design has higher ergodic capacity than the punctured Perfect code. Simulation results which corroborate these claims are presented.

Decorrelation of broad band signals by toeplitz-block-toeplitz structuring

We propose F-norm of the cross-correlation part of the array covariance matrix as a measure of correlation between the impinging signals and study the performance of different decorrelation methods in the broadband case using this measure. We first show that dimensionality of the composite signal subspace, defined as the number of significant eigenvectors of the source sample covariance matrix, collapses in the presence of multipath and the spatial smoothing recovers this dimensionality. Using an upper bound on the proposed measure, we then study the decorrelation of the broadband signals with spatial smoothing and the effect of spacing and directions of the sources on the rate of decorrelation with progressive smoothing. Next, we introduce a weighted smoothing method based on Toeplitz-block-Toeplitz (TBT) structuring of the data covariance matrix which decorrelates the signals much faster than the spatial smoothing. Computer simulations are included to demonstrate the performance of the two methods.

High-rate, Double-Symbol-Decodable STBCs from Clifford Algebras

For the number of transmit antennas N = 2(a) the maximum rate (in complex symbols per channel use) of all the Quasi-Orthogonal Designs (QODs) reported in the literature is a/2(a)-1. In this paper, we report double-symbol-decodable Space-Time Block Codes with rate a-1/2(a)-2 for N = 2(a) transmit antennas. In particular, our code for 8 and 16 transmit antennas offer rates 1 and 3/4 respectively, the known QODs offer only 3/4 and 1/2 respectively. Our construction is based on the representations of Clifford algebras and applicable for any number of transmit antennas. We study the diversity sum and diversity product of our codes. We show that our diversity sum is larger than that of all known QODs and hence our codes perform better than the comparable QODs at low SNRs for identical spectral efficiency. We provide simulation results for various spectral efficiencies.