Column Direction Research Articles

Median filtering detection using variation of neighboring line pairs for image forensics

Attention to tampering by median filtering (MF) has recently increased in digital image forensics. For the MF detection (MFD), this paper presents a feature vector that is extracted from two kinds of variations between the neighboring line pairs: the row and column directions. Of these variations in the proposed method, one is defined by a gradient difference of the intensity values between the neighboring line pairs, and the other is defined by a coefficient difference of the Fourier transform (FT) between the neighboring line pairs. Subsequently, the constructed 19-dimensional feature vector is composed of these two parts. One is the extracted 9-dimensional from the space domain of an image and the other is the 10-dimensional from the frequency domain of an image. The feature vector is trained in a support vector machine classifier for MFD in the altered images. As a result, in the measured performances of the experimental items, the area under the receiver operating characteristic curve (AUC, ROC) by the sensitivity (PTP: the true positive rate) and 1-specificity (PFP: the false-positive rate) are above 0.985 and the classification ratios are also above 0.979. Pe (a minimal average decision error) ranges from 0 to 0.024, and PTP at PFP=0.01 ranges from 0.965 to 0.996. It is confirmed that the grade evaluation of the proposed variation-based MF detection method is rated as “Excellent (A)” by AUC is above 0.9.

Dense Projection Tomography on the Triangular Tiling

In this paper, we consider the binary tomography reconstruction problem. A new ap- proach is proposed what exploits a possibility provided by the natural structure of the triangular grid, which is not available in the case of the classical square grid. In contrast to the square grid, in the case of the triangular grid information need for the reconstruction of the unknown image is increasing when not only one, but two projections are used by lanes. In this way, the number of 4 andr shaped pixels per lane can be determined. We propose this type of projection approach and call it dense projections. The reconstruction is based on three projection directions by the lane directions of the grid (they are analogous to row and column directions on the square grid). Our algorithm is deterministic and uses energy minimization technique to find (near) optimal solution in a reasonable time. The experimental evaluation of the new method, using regular hexagon shaped test images, is given. Comparison with reconstructions based on the square grid is also considered.

Hexagonal packing lattice formed by functionalized gold nanoparticles

Nanostructures formed by functionalized gold nanoparticles (GNPs) were studied using X-ray diffraction (XRD). Two systems of such functionalized GNPs, one with gold core grafted by both ligand (rod-like liquid crystalline molecule) and co-ligand (alkylthiols) and the other one with only ligand grafted, were studied. The presence of ligands and co-ligands increases the mobility of GNPs and induces long-range ordered packing style with the change of temperature. In both systems, functionalized GNPs form columns first and then pack on hexagonal lattice. It is found that the correlation along column direction is dependent on surface conditions of GNPs, which is observed for system with only ligand grafted. Based on experimental results, size of gold core and number of ligand and co-ligand belonging to each gold core were calculated. Diffraction intensities and electron density map of each structure were simulated, which are evidently similar to experimental ones. Finally, structure model was proposed and discussed. The results help us to understand and control the way that organic molecules affect the organization of nanoparticles, and facilitate the design of new hybrid materials. Gold nanoparticles (yellow) functionalized by liquid crystalline ligand (green) and co-ligand (red) can form ordered 3D hexagonal columnar phase. In this structure, ligands congregate around the equator of the gold nanoparticles, while co-ligands aggregate near the north and south poles. This enables the nanoparticles to align in strings wrapped in the sheaths of uniaxially oriented mesogens forming columns, and then these columns are packed on a hexagonal lattice.

Field experimental design comparisons to detect field effects associated with agronomic traits in upland cotton

Field variation is one of the important factors that can have a significant impact on genetic data analysis. Ineffective control of field variation may result in an inflated residual variance and/or biased estimation of genetic variations and/or effects. In this study, we addressed this problem by merging genetic models with the information from a rectangular cotton field layout (referred to row and column directions). Data from a genetic mapping study in Upland cotton (Gossypium hirsutum L.) was used to validate the proposed methodology. This study included model evaluation based on simulations and actual data analysis on four agronomic traits (seed yield, lint yield, lint percentage, and boll weight) in cotton. Results based on simulations suggested that when there were no row and column effects, the conventional and the extended genetic models yielded similar results. However, when either field row and/or column effects were significant, the conventional genetic model yielded biased estimates for residual variance component with larger mean square error whereas the extended genetic models yielded more unbiased estimates. Actual data analysis revealed that lint yield and seed yield were significantly influenced by the systematic variation present in the field. With the extended model, the residual variance associated with these traits was reduced approximately 65 % compared to the conventional block model. Accordingly, the averaged heritability estimate increased by about 18 % for these traits. Thus, the results suggested that genetic data analysis can be improved when field variation is considered.

Bilinear low-rank coding framework and extension for robust image recovery and feature representation

We mainly study the low-rank image recovery problem by proposing a bilinear low-rank coding framework called Tensor Low-Rank Representation. For enhanced low-rank recovery and error correction, our method constructs a low-rank tensor subspace to reconstruct given images along row and column directions simultaneously by computing two low-rank matrices alternately from a nuclear norm minimization problem, so both column and row information of data can be effectively preserved. Our bilinear approach seamlessly integrates the low-rank coding and dictionary learning into a unified framework. Thus, our formulation can be treated as enhanced Inductive Robust Principal Component Analysis with noise removed by low-rank representation, and can also be considered as the enhanced low-rank representation with a clean informative dictionary via low-rank embedding. To enable our method to include outside images, the out-of-sample extension is also presented by regularizing the model to correlate image features with the low-rank recovery of the images. Comparison with other criteria shows that our model exhibits stronger robustness and enhanced performance. We also use the outputted bilinear low-rank codes for feature learning. Two unsupervised local and global low-rank subspace learning methods are proposed for extracting image features for classification. Simulations verified the validity of our techniques for image recovery, representation and classification.

Power IC용 저면적 Differential Paired eFuse OTP 메모리 설계

본 논문에서는 Power IC용 저면적 32비트 differential paired eFuse OTP 메모리를 설계하였다. OTP 메모리 셀 어레이에서 행의 개수가 열의 개수보다 더 작은 경우 eFuse 링크의 프로그램 전류를 공급하는 SL (Source Line) 구동 라인을 열 방향으로 라우팅하는 대신 행 방향으로 라우팅하므로 레이아웃 면적을 많이 차지하는 SL 구동회로의 수를 줄이는 differential paired eFuse 셀 어레이 방식과 코어 회로를 제안하였다. 그리고 blowing되지 않은 eFuse 링크가 EM (Electro-Migration) 현상에 의해 blowing되는 불량을 해결하기 위해 RWL (Read Word-Line) 구동 회로와 BL (Bit-Line) 풀-업 부하회로에 V2V (<TEX>$=2V{\pm}0.2V$</TEX>)의 regulation된 전압을 사용하였다. 설계된 32비트 eFuse OTP 메모리의 레이아웃 면적은 <TEX>$228.525{\mu}m{\times}105.435{\mu}m$</TEX>으로 기존의 셀 어레이 라우팅을 이용한 IP 크기인 <TEX>$197.485{\mu}m{\times}153.715{\mu}m$</TEX> 보다 20.7% 더 작은 것을 확인하였다. In this paper, a small-area 32-bit differential paired eFuse OTP memory for power ICs is designed. In case of smaller number of rows than that of columns for the OTP memory cell array, a scheme for the cell array reducing the number of SL driver circuits requiring their larger layout areas by routing the SL (source line) lines supplying programming currents for eFuse links in the row direction instead of the column direction as well as a core circuit is proposed. In addition, to solve a failure of being blown for non-blown eFuse links by the electro-migration phenomenon, a regulated voltage of V2V (<TEX>$=2V{\pm}0.2V$</TEX>) is used to a RWL (read word line) driver circuit and a BL (bit line) pull-up driver circuit. The layout size of the designed 32-bit eFuse OTP memory is <TEX>$228.525{\mu}m{\times}105.435{\mu}m$</TEX>, which is confirmed to be 20.7% smaller than that of the counterpart using the conventional cell array routing, namely <TEX>$197.485{\mu}m{\times}153.715{\mu}m$</TEX>.

Three-dimensional heat transfer coefficient distributions in a large horizontal-tube falling film evaporator

The distributions of the heat transfer coefficient in a large horizontal-tube falling film evaporator under different operating conditions were investigated with numerical simulation in this paper based on the distributed parameter model (DPM). The heat transfer coefficient distributions along the tube row direction, tube length direction, and tube column direction were analyzed with the brine inlet spray density ranging from 0.05kg/ms to 0.09kg/ms and the brine inlet salinity from 30g/kg to 50g/kg. Results show that the heat transfer coefficient has higher values for the evaporating zone than the preheating zone. With the operating conditions included in this paper, the change of brine inlet salinity leads to a wider variation range of heat transfer coefficient along the tube length direction for the evaporating zone than the change of brine inlet spray density. The change of brine inlet spray density leads to a wider variation range of heat transfer coefficient along the tube row direction for the first tube pass and along the tube length direction for the preheating zone than the change of brine inlet salinity.

Characteristic study of steam maldistribution in horizontal-tube falling film evaporators

The maldistribution of steam inlet velocity of a large horizontal-tube falling film evaporator was investigated with numerical simulation using the distributed parameter model (DPM) to provide essential design of falling film evaporator. Empirical correlations were obtained through experimental studies on the falling film evaporation outside a single horizontal tube, the condensation inside the tube and the inter-tube vapor flow across a tube bundle to conduct the three-dimensional simulation of the evaporator. The maldistribution of tube-side steam on tube row and column directions were analyzed with the brine inlet salinity ranging from 30 g/kg to 60 g/kg and the brine inlet spray density from 0.05 kg/m·s to 0.09 kg/m s. The Brine boiling point elevation (BPE) and the inter-tube vapor flow resistance were taken into account in the model. Considering the self-compensation characteristic between the heat flux and steam inlet velocity of the tubes, the maldistribution of heat flux was presented to analyze the maldistribution of tube-side steam inlet velocity. The simulation results exhibit the decreasing trend of the tube-side steam inlet velocity non-uniformity along the tube row direction with the increment in the brine inlet spray density and the decrement in the brine inlet salinity. Along the tube column direction, the non-uniformity of steam inlet velocity increases with the increment of brine inlet spray density and the decrement of brine inlet salinity.

An image encryption scheme using generalized Arnold map and affine cipher

This paper aims to provide an image encryption scheme with an efficient bit-level permutation and a pixel-level diffusion procedure. In the bit-level permutation, we divide each pixel into 8 bits, and arrange the positions of each bit by the generalized Arnold map in row and column direction. Hence, a significant diffusion effect is happened in the bit-level permutation. In the pixel-level diffusion procedure, we apply affine cipher to change the gray value and the histogram distribution of the permutated image. Various types of security analyses demonstrate that the proposed scheme is competitive with that ordinary permutation–diffusion type image cipher and proper for practical image encryption.

Line Scan Sensor용 저면적 eFuse OTP 설계

In this paper, a small-area cell array method of reducing number of SL drivers requiring large layout areas, where the SL drivers supplying programming currents are routed in the row direction in stead of the column direction for eFuse OTP memory IPs having less number of rows than that of columns such as a cell array of four rows by eight columns, and a core circuit are proposed. By adopting the proposed cell array and core circuit, the layout area of designed 32-bit eFuse OTP memory IP is reduced. Also, a V2V (=2V±10％) regulator necessary for RWL driver and BL pull-up load to prevent non-blown eFuse from being blown from the EM phenomenon by a big current is designed. The layout size of the designed 32-bit OTP memory IP having a cell array of four rows by eight columns is 13.4% smaller with 120.1㎛ × 127.51㎛ (=0.01531㎜ 2 ) than that of the conventional design with 187.065㎛ × 94.525㎛ (=0.01768㎜ 2 ).

Evaluation of dispersal volcanic products of recent events in lichens in environmental gradient, Nahuel Huapi National Park, Argentina.

The atmospheric transport of volcanic products are subject to several variables, mainly the height of the eruption column and wind direction, thus elements associated with the ashes are deposited in major or lesser degree depending on variables as latitude, wind and humidity. The lichens are able to reflect the atmospheric fallout. The present work evaluated the correlation between meteorological parameters, geographic locations, sulphur and other element concentrations in lichens genus Usnea affected by Puyehue-Cordón Caulle complex (North Patagonia Andean Range) eruption of June 4, 2011. Semiquantitative analyses of biological elements by scanning electron microscope methods, sulphur (S) by LECO and other elements by instrumental neutron activation were evaluated by principal component analysis. Elements as antimony, arsenic, barium, bromine, calcium, caesium, potassium, rubidium, selenium, and uranium correlated with distance to volcano, also calcium and potassium with longitude while bromine, rubidium, and potassium with humidity. Those results indicate that Usnea sp. is a good bioindicator of the atmospheric volcanic emissions in relation to environmental gradient.

Palm vein recognition based on directional empirical mode decomposition

Directional empirical mode decomposition (DEMD) has recently been proposed to make empirical mode decomposition suitable for the processing of texture analysis. Using DEMD, samples are decomposed into a series of images, referred to as two-dimensional intrinsic mode functions (2-D IMFs), from finer to large scale. A DEMD-based ð2-DÞ 2 linear discriminant analysis (LDA) for palm vein recognition is proposed. The pro- posed method progresses through three steps: (i) a set of 2-D IMF features of various scale and orientation are extracted using DEMD, (ii) the ð2-DÞ 2 LDA method is then applied to reduce the dimensionality of the feature space in both the row and column directions, and (iii) the nearest neighbor classifier is used for classification. We also propose two strategies for using the set of 2-D IMF features: ensemble DEMD vein representation (EDVR) and multichannel DEMD vein representation (MDVR). In experiments using palm vein databases, the proposed MDVR-based ð2-DÞ 2 LDA method achieved recognition accuracy of 99.73%, thereby demonstrating its feasibility

A 2DPCA Based Semi-Supervised Learning and its Application on Face Recognition

A Mahalanobis distance based semi-supervised fuzzy clustering model is presented in this paper, whose objective function has a good explanation on how the labeled and unlabeled data are used in finding the underlying structure of matrix data. The iterative algorithm to solve this model is given. This algorithm can directly deal with matrix data like face images. We use 2DPCA on both row and column directions to reduce the dimension of image faces. The experimental result shows that using 2DPCA and semi-supervised algorithms can have a fairly good recognition rate if enough labeled data are given.

Neighbor Designs: A New Approach of Local Control

It is well known that randomization, replication and local control play important role in experimental design. Neighbor designs provide a tool for local control in situation where experimental units are influenced by neighboring units. A neighbor design is called one-dimensional if neighbor effects are controlled in only one way, i.e., either in row or in column direction. In two-dimensional design, neighbor effects are controlled in both ways (rows and columns). In this paper the concept of neighbor designs, its types and importance is discussed with examples. Models of Neighbor effects for different situations are also discussed.

Estimation of linear transformation by analyzing the periodicity of interpolation

Linear transformation, such as rotation, scaling, or any combinations of geometric attacks, is among the most common forms of image manipulation. This letter proposes a forensic technique that estimates the linear transformation of an investigated image. We exploited the periodic properties of interpolation by the second-derivative of the transformed image in both the row and column directions. Both the magnitude and phase information of the derived signals were analyzed to estimate the transformation matrix accurately. Empirical evidence from a large database of manipulated images indicates the superior performance of the proposed method.

Odd or Even? Monolayer Domain Size Depends on Diyne Position in Alkadiynylanthracenes

1,5-(Alkadiynyl)anthracenes self-assemble single component and multicomponent monolayers at the solution-HOPG interface. An alkadiynyl chain's kinked shape constrains the molecular structures with which it can close-pack. This affords rudimentary molecular recognition that has been used to direct self-assembly of 1-D patterned, multicomponent monolayers. The unit cell building blocks of single- and multicomponent alkadiynylanthracene monolayers repeat with high fidelity for 100s of nanometers along the side chain direction. Unit cell repeat fidelity along the orthogonal, anthracene column direction of the monolayer depends on diyne location within the side chain; even-position diyne side chains produce high fidelity of unit cell repeats and wider domain widths along the anthracene columns, whereas odd-position diyne side chains produce more frequent domain interfaces that disrupt the anthracene columns. Alkadiynylanthracene monolayers may be viewed as stacks of 1-D molecular tapes. 1-D tape molecular composition, sequence, and intratape side chain alignment are dictated by shape complementarity of the kinked alkadiynyl side chains. Stacking alignments of adjacent 1-D tapes are controlled by shape matching of tape peripheries and determine repeat fidelity along the anthracene columns. Tapes stacked with a constant intertape alignment comprise crystalline domains that repeat along the anthracene columns. The 1-D tapes formed by anthracenes with odd-position diynes have triangle wave peripheries that close-pack in multiple stacking alignments. This reduces unit cell repeat fidelity and decreases the widths of crystalline domains along the anthracene columns. Even-position diyne side chains form 1-D tapes with trapezoid wave peripheries that close-pack in only one stacking alignment. This generates higher stacking fidelity, larger domain widths, and fewer domain interfaces along the anthracene columns of even-position diyne monolayers. Even- and odd-position diyne monolayers exhibit comparable densities of interfaces between enantiotopic domains and between domains aligned along different graphite symmetry axes. These interfaces likely arise through collisions of independently nucleated/growing domains and persist for lack of kinetically competent pathways that interconvert or merge the domains.

Bilateral diagonal principal component analysis based on matrix distance metrics: a projection technique for face identification

Abstract. One of the most recent techniques developed for face rec-ognition is diagonal principal component analysis (DiaPCA). In con-trast to two-dimensionalPCA (2DPCA), DiaPCAis basedon diagonalrepresentation of face images. Therefore, the correlations betweenvariations of rows and those of columns can be kept. This paper pro-poses a novel projection scheme called bilateral DiaPCA (BDiaPCA)which simultaneously performs left and right projections of originalfaceimagesin anew subspace derivedfromdiagonalrepresentation.Therefore BDiaPCA reduces the dimension of the original imagematrix in both column and row directions. Experiments were carriedout under varying illumination and facial expression using three well-known face databases: Yale, PF01, and a subset of FERET contain-ing 200 people. Comparisons with 2DPCA and DiaPCA show thatBDiaPCA outperforms these methods and achieves higher recogni-tion rates using the Yang, Frobenius, or assembled matrix distancemetrics. © 2013 SPIE and IS&T [DOI: 10.1117/1.JEI.22.2.023020]

Two-directional two-dimensional modified Fisher principal component analysis: an efficient approach for thermal face verification

In recent years, verification based on thermal face images has been extensively studied because of its invariance to illumination and immunity to forgery. However, most of them have not given full consideration to high-verification performance and singular within-class scatter matrix problems. We propose a novel thermal face verification algorithm, which is named two-directional two-dimensional modified Fisher principal component analysis. First, two-dimensional principal component analysis (2-DPCA) is utilized to extract the optimal projective vector in the row direction. Then, 2-D modified Fisher linear discriminant analysis is implemented to overcome the singular within-class scatter matrix problem of the 2-DPCA space in the column direction. Comparative experiments on the natural visible and infrared facial expression thermal face subdatabase demonstrate that the proposed approach outperforms state-of-the-art methods in terms of verification performance.

2D2LPI

In this paper, a new method of representing images called two directional two dimensional locality preserving indexing called 2D2LPI is presented. It is an extension of the two dimensional locality preserving indexing (2DLPI) method. The authors argue that the recently proposed 2DLPI reduces the dimensions of images in row direction and we propose an alternate way of reducing the dimension in column direction. Later the authors propose a method to reduce the size of an image both in row and column directions. To corroborate the efficacy of the proposed two directional two dimensional approach the authors design a model for person identification based on single instance of finger knuckle print and subsequently the authors propose a feature level fusion of multi-instance finger knuckle print for person identification. Also to study the suitability of the proposed approach on a different domain, a study on video summarization is also presented in this paper. The results of the proposed method are compared with that of the state of the art techniques such as 2D2PCA, 2D2LPP and it is found that the proposed 2D2LPI model is more competitive in terms of accuracy.

Palm vein recognition based on a modified $$\text{(2D)}^{2}\text{ LDA}$$

Biometric computing offers an effective approach to personal identification based on unique, stable physical or behavioral characteristics. A reliable and robust personal verification approach using palm vein patterns is presented in this paper. This approach has lower computational and memory requirements and a higher recognition accuracy than similar methods. In my work, a near-infrared charge-coupled device camera is adopted as an input device for capturing palm vein images, since it can provide low-cost, non-contact imaging. In the proposed approach, two finger-webs are automatically selected to define the region of interest (ROI) in the palm vein images. Modified two-directional two-dimensional linear discriminant analysis (\(\text{(2D)}^{2}\text{ LDA}\)), which performs an alternate two-dimensional LDA (2DLDA) in the column direction of images in the 2DLDA subspace, is proposed to exploit the correlation between rows and columns of palm vein features inside the ROI. The major advantage of the method is that it requires fewer coefficients for efficient palm vein image representation and recognition. A total of 4,140 palm vein images were collected from 207 persons to verify the validity of the proposed palm vein recognition approach. High accuracies (\(>\)99 %) have been obtained by the proposed method, and the speed of the method (response time \(<\)0.75 s) is rapid enough for real-time recognition. Experimental results demonstrate that the proposed approach is feasible and effective for palm vein recognition.