Low Computational Cost Method Research Articles

Superoxide Radical Anion Adduct of 5,5-Dimethyl-1-pyrroline N-Oxide. 4. Conformational Effects on the EPR Hyperfine Splitting Constants

Spin trapping has been commonly employed in the detection of superoxide radical anion in chemical and biological systems; hence, accurate interpretation of the hyperfine splitting constants (hfsc's) arising from the O(2)(*-) adducts (also referred to as hydroperoxyl (HO(2)(*)) radical adducts) of various nitrones is important. In this work, the nature of the relevant hfsc's was investigated by examining the effect of conformational changes in the hydroperoxyl moiety of the O(2)(*-) adducts of 5,5-dimethyl-1-pyrroline N-oxide (DMPO), 5-ethoxycarbonyl-5-methyl-1-pyrroline N-oxide (EMPO), 5-diethoxyphosphoryl-5-methyl-1-pyrroline N-oxide (DEPMPO), 5-carbamoyl-5-methyl-1-pyrroline N-oxide (AMPO), and 7-oxa-1-azaspiro[4.4]non-1-en-6-one N-oxide, (CPCOMPO) on the magnitude of a(N), a(beta-H), and a(gamma-H). Conformational change around the substituents and their effect on the hfsc's were also explored. Results indicate that a(beta-H) is most sensitive to conformational changes of the hydroperoxyl and substituent groups relative to hfsc's of other nuclei. The orbital overlap between the C-H sigma-orbital and the SOMO of the nitroxyl nitrogen plays a crucial factor in determining the magnitude of the a(beta-H). The hfsc values for the O(2)(*-) adducts were predicted with high accuracy by using a low-cost computational method at the PCM(water)/BHandHLYP/EPR-III//B3LYP/6-31G* level of theory without taking into account the explicit water interaction.

Calculation of the enthalpies of formation and proton affinities of some isoquinoline derivatives

Ab initio molecular orbital theory has been used to calculate enthalpies of formation of isoquinoline, 1-hydroxyisoquinoline, 5-hydroxyisoquinoline, and 1,5-dihydroxyisoquinoline as well as some pyridine and quinoline derivatives. The proton affinities of the four isoquinoline derivatives were also obtained. The high-level composite methods G3(MP2), G3(MP2)//B3LYP, G3//B3LYP, and CBS-QB3 have been used for this study, and the results have been compared with available experimental values. For six of the eight studied compounds, the theoretical enthalpies of formation were very close to the experimental values (to within 4.3 kJ · mol −1); where comparison was possible, the theoretical and experimental proton affinities were also in excellent agreement with one another. However, there is an extraordinary discrepancy between theory and experiment for the enthalpies of formation of 1-hydroxyisoquinoline and 1,5-dihydroxyisoquinoline, suggesting that the experimental values for these two compounds should perhaps be re-examined. We also show that popular low cost computational methods such as B3LYP and MP2 show very large deviations from the benchmark values.

Automated Handwritten Characters Recognition Based on a Vector Field

In order to obtain a low computational cost method for automatic handwritten characters recognition, this paper proposes a combined system of two rough classification methods based on features of a vector field: one is an autocorrelation matrix method, and another is a low frequency Fourier expansion method. In each method, the similarity is defined as a weighted sum of the squared values of the inner product between input pattern and the reference patterns that are normalized eigenvectors of KL (Karhunen-Loeve) expansion. This paper also describes a way of deciding the weight coefficients using Linear Regression Method, and shows the effectiveness of the proposed method by illustrating some experimentation results for 3036 categories of handwritten Japanese characters.

A weighted linear prediction method for near-field source localization

This paper proposes a low computational cost method for the near-field narrowband source localization problem, which does not require multidimensional search or high-order statistics. The proposed method is based on the second-order statistics (SOS) of the outputs of a uniform linear array (ULA). More precisely, the range and angle parameters are estimated through a weighted linear prediction (LP) algorithm applied to a properly chosen array output correlation sequence. Detailed performance analysis and derivation of the optimal weightings are provided. Simulation results are finally presented to validate the theoretical analysis results and to assess the performance of the proposed method.

Resonance Raman spectra simulation of the biphenyl anion and cation radicals

A low-cost computational method, based on the restricted active space CASSCF procedure, is used to simulate resonance Raman spectra of the biphenyl anion and cation radicals. Theoretical and experimental spectra are in good agreement for both compounds, for the perhydrogenated and perdeuterated derivatives. These calculations confirm that the anion has a planar structure, and encourage concluding to a twisted conformation for the cation.

Efficient multispectral texture segmentation using multivariate statistics

A complete, low computational cost method is presented for multispectral textured image segmentation. The procedure performs a tesselation of the image into non-overlapped rectangular regions and decides about the homogeneity of each region, using statistical hypothesis testing. Regions labelled as homogeneous are used to estimate the parameters that are necessary to classify the pixels of the heterogeneous regions. The proposed scheme can also be used to estimate the number of different textures in the image. This represents an efficient alternative to other computationally expensive methods, such as those that employ clustering techniques.

Characters Recognition Method Based on Vector Field and Simple Linear Regression Model

In order to obtain a low computational cost method (or rough classification) for automatic handwritten characters recognition, this paper proposes a combined system of two feature representation methods based on a vector field: one is autocorrelation matrix, and another is a low frequency Fourier expansion. In each method, the similarity is defined as a weighted sum of the squared values of the inner product between input pattern feature vector and the reference pattern ones that are normalized eigenvectors of KL (Karhunen-Loeve) expansion. This paper also describes a way of deciding the weight coefficients using a simple linear regression model, and shows the effectiveness of the proposed method by illustrating some experimentation results for 3036 categories of handwritten Japanese characters.