Statistical analysis of human visual impressions on morphological image manipulation of gray scale textures

A method of evaluating human visual impressions of gray scale textures using morphological manipulation is proposed. To study the effects of textural features on human Kansei, we introduced a texture analysis method based on mathematical morphology. Kansei is a Japanese word for sensibility or emotion. Kansei engineering is an approach to connect human sensibility with engineering applications. The proposed method allows us to manipulate global and local properties of a texture separately. Variations of textures were generated by repetitively modifying arranged objects and configurations of the arrangements of original textures. The manipulated textures were presented to human respondents and the similarity of those textures based on human impressions was evaluated. Hierarchical clustering was applied to the similarity matrix generated from respondents’ observations. The results of the human evaluation were compared with that of the objective similarity evaluation adopting six global textural features. The global features such as density, regularity, and directionality of the point configurations were shown to have significant effects on human visual impressions and identification of textures. In the case of a texture without significant characteristics in its point configuration, local features such as grain shape have an effect on visual impressions.     

真实加工表面后向散射特性分析及表面参数反演

无合作目标激光绝对测距中,远距离、真实表面的后向散射特性机理不明朗,严重影响测距结果,是制约测距技术发展的重要瓶颈。以立铣、平铣、平磨等三种典型实际机械加工方式下的粗糙表面为研究对象,测量了在1 550 nm红外激光照射下形成的后向散射场,探究了不同加工方式下特殊后向散射场形成的原因,深入分析了表面纹理、入射方位角、入射角度、粗糙度对后向散射场分布的影响。实验结果分析表明,加工制造表面的后向散射光谱形态分布受加工方式的影响很大,且相互入射几何关系和粗糙度对每种加工方式下的实际粗糙表面均有规律性影响。为了能够获取足够的后向散射能量,对表面参数的识别反演显得十分重要。进一步构建了一种加工表面多维参数反演模型,采用另外一种加工方式(刨床)的样块数据进行验证,加工方式能够被准确区分,入射方位角和粗糙度反演的相对误差分别达到1.21%和1.03%,反演精度较高。经实验验证,通过表面参数的反演极大拓宽了无合作目标激光绝对测距的范围,有效降低了表面纹理、入射方位角、粗糙度等对测距范围的影响。这一研究结果还对具有纹理特征加工表面的后向散射光谱的研究和在其他领域的应用均具有一定的参考价值。     

Anomalous scaling for thick electrodeposited films

Anomalous surface roughness scaling, where both the local and the large-scale roughness show a power-law dependence on the film thickness, has been widely observed. Here we show that the value of the local roughness exponent in the early stages of Cu electrodeposition depends on the deposition potential. However, initial anomalous scaling can lead to two qualitatively different types of behavior for large film thickness (t>/ or =4 microm). For Cu films electrodeposited with forced convection at high potential and current density, the anomalous scaling is transient: the local roughness saturates for the thickest films studied. When Cu films are electrodeposited at similar potential and current density but with reduced convection, no saturation of the local roughness is observed. Instead the film forms overhangs such that the surface height becomes a multivalued function of the lateral position.     

基于激光散斑成像的零件表面粗糙度建模

表面粗糙度是衡量机械表面加工水平的重要参数. 通过构建一套激光散斑成像采集系统, 获取了不同表面加工类型和不同粗糙度值的零件表面激光散斑图像. 应用Tamura纹理特征理论提取图像的纹理粗糙度、对比度、方向度特征, 并分析了这三个特征与表面粗糙度的关系. 发现了纹理粗糙度特征与表面粗糙度的单调关系, 推导出平磨、外磨、研磨三种表面加工工艺的粗糙度值与图像纹理粗糙度特征的数学函数关系, 实现了表面粗糙度的测量. 同时, 利用Tamura纹理特征与加工工艺的依赖关系, 建立了基于贝叶斯网络的工艺识别推理模型, 推理出了零件表面加工工艺. 通过为多种加工类型表面建立粗糙度测量模型, 为粗糙度测量提供了新思路. 实验证明所提的粗糙度测量模型能以较高的准确率识别出零件表面加工类型并测量出其表面粗糙度值.     

基于Contourlet变换与FLD的人脸表情识别方法

提出了一种基于Contourlet变换与FLD的人脸表情特征提取方法。Contourlet变换是一种新的多尺度几何分析方法,它在具备小波变换的多分辨率特性和时频局部特性的同时,还具有很强的多方向选择性和各向异性。图像经过Contourlet变换后的低频分量可体现表情的概貌,高频方向子带可体现表情的轮廓和纹理等细节。将低频分量与部分高频方向子带结合起来作为整体特征,可压缩图像的数据量,并能够体现表情的本质特征。首先用Fisher线性判别法(FLD)进行特征提取,然后采用K-近邻法进行分类。实验证明,该方法比经典的FLD方法有着更快的特征提取速度和更高的表情识别率。     

Acceleration of the iterative solver in the discrete dipole approximation: Application to the orientation variation of irregularly shaped particles

We have applied a method of reducing the number of iterations required to solve a system of linear equations in the discrete dipole approximation. This method obtains an initial guess of dipole polarization from those with similar particle characteristics (e.g., the size parameter and refractive index) calculated a priori. If the initial guess is closer to the solution, the number of iterations of the linear equation solution becomes smaller than that calculated with an arbitrary initial value.

This method was applied to various particle orientations using spline interpolation of the initial guess of dipole polarization from orientations calculated a priori.

We studied three types of particle model: an aggregate, a deformed sphere with moderate surface roughness, and a particle with a large number of edges. For the particle with a large number of edges, we propose a new model called the overlapping mixture of multiple tetrahedra (OMMT).

The proposed method is most advantageous for particles with moderate surface roughness (e.g., a deformed sphere), for which the calculation time was reduced to 20–40% of the original calculation time. For OMMT and an aggregate, the computation time was reduced to 30–60% and 40–90%, respectively. The differences in the scattering coefficient, absorption coefficient, intensity and polarization introduced by our method were less than 0.008%, 0.03%, 0.1%, and 0.08%, respectively.

If the light scattering properties vary slowly with the orientation variation, interpolation of the results is more efficient than the proposed method and produces only a small difference in the results. However, the interpolation of the results fails for particles such as BCCA64, for which our proposed method produces more accurate results.     

一种海洋环境噪声分步反演地声参数方法

分析不同海底参数对环境噪声垂向空间特性的敏感度,根据海底密度对环境噪声垂直指向性的小掠射角部分不敏感,而对等效海底损失的大掠射角部分相对敏感的特点,提出了一种海洋环境噪声分步反演地声参数方法:先用环境噪声垂直指向性小掠射角部分反演海底声速、衰减;之后利用大掠射角部分来反演海底密度。仿真算例和海上实验数据处理结果验证了该方法的有效性。结果表明:分步反演加强了匹配物理量对海底参数敏感部分的关注,在反演结果精度(海底密度)以及反演效率上都有所改善,具有一定可行性。      

一种适于空间多光谱TDICCD图像压缩算法

针对传统遥感图像压缩算法中小波变换和位平面编码没有考虑图像内容特点而导致多光谱各谱段图像边缘和纹理的模糊的问题,提出一种适于成像谱段数相对较少多光谱TDICCD图像压缩算法。提出的自适应提升DWT可以自适应的选择最佳的提升方向,同时根据图像局部特征使用拉格朗日插值策略进行预测,这种方法可以充分的利用图像的纹理信息。提出的码率控制算法可以根据图像纹理复杂程度进行自适应的码率分配。实验结果表明,提出的压缩算法具有良好压缩性能,对于平滑的图像与传统算法相当,对于边缘和纹理程度复杂的图像,性能高于传统算法,在正常工作压缩比为8∶1时,平均信噪比比传统方法平均提高了3.53 dB。有效的保护了多光谱图像的边缘和纹理信息,非常适于纹理复杂的空间多光谱CCD图像压缩应用。     

Rapid determination of surface roughness of polycrystalline silicon following frontside and backside excimer laser irradiation

Excimer-laser crystallization (ELC) is the most commonly employed technology for fabricating low-temperature polycrystalline silicon (LTPS). Investigations on the surface roughness of polycrystalline silicon (poly-Si) thin films have become an important issue because the surface roughness of poly-Si thin films is widely believed to be related to its electrical characteristics. In this study, we develop a simple optical measurement system for rapid surface roughness measurements of poly-Si thin films fabricated by frontside ELC and backside ELC. We find that the incident angle of 20° is a good candidate for measuring the surface roughness of poly-Si thin films. The surface roughness of polycrystalline silicon thin films can be determined rapidly from the reflected peak power density measured by the optical system developed using the prediction equation. The maximum measurement error rate of the optical measurement system developed is less than 9.71%. The savings in measurement time of the surface roughness of poly-Si thin films is up to 83%. The method of backside ELC is suggested for batch production of low-temperature polycrystalline silicon thin-film transistors due to the lower surface roughness of poly-Si films and higher laser-beam utilization efficiency.     

An improved fast fractal image compression using spatial texture correlation

This paper utilizes a spatial texture correlation and the intelligent classification algorithm (ICA) search strategy to speed up the encoding process and improve the bit rate for fractal image compression. Texture features is one of the most important properties for the representation of an image. Entropy and maximum entry from co-occurrence matrices are used for representing texture features in an image. For a range block, concerned domain blocks of neighbouring range blocks with similar texture features can be searched. In addition, domain blocks with similar texture features are searched in the ICA search process. Experiments show that in comparison with some typical methods, the proposed algorithm significantly speeds up the encoding process and achieves a higher compression ratio, with a slight diminution in the quality of the reconstructed image; in comparison with a spatial correlation scheme, the proposed scheme spends much less encoding time while the compression ratio and the quality of the reconstructed image are almost the same.     

Fractal analysis and atomic force microscopy measurements of surface roughness for Hastelloy C276 substrates and amorphous alumina buffer layers in coated conductors

In coated conductors, surface roughness of metallic substrates and buffer layers could significantly affect the texture of subsequently deposited buffer layers and the critical current density of superconductor layer. Atomic force microscopy (AFM) is usually utilized to measure surface roughness. However, the roughness values are actually relevant to scan scale. Fractal geometry could be exerted to analyze the scaling performance of surface roughness. In this study, four samples were prepared, which were electro polished Hastelloy C276 substrate, mechanically polished Hastelloy C276 substrate and the amorphous alumina buffer layers deposited on both the substrates by ion beam deposition. The surface roughness, described by root mean squared (RMS) and arithmetic average (R_a) values, was analyzed considering the scan scale of AFM measurements. The surfaces of amorphous alumina layers were found to be fractal in nature because of the scaling performance of roughness, while the surfaces of Hastelloy substrates were not. The flatten modification of AFM images was discussed. And the calculation of surface roughness in smaller parts divided from the whole AFM images was studied, compared with the results of actual AFM measurements of the same scan scales.     

An equivalent roughness model for seabed backscattering at very high frequencies using a band-matrix approach

This work concerns modeling of very high frequency (>100 kHz) sonar images obtained from a sandy seabed. The seabed is divided into a discrete number of ID height profiles. For each height profile the backscattered pressure is computed by an integral equation method for interface scattering between two homogeneous media as formulated by Chan [IEEE Trans. Antennas Propag. 46, 142-149 (1998)]. However, the seabed is inhomogeneous, and volume scattering is a major contributor to backscattering. The SAX99 experiments revealed that the density in the unconsolidated sediment within the first 5 mm exhibits a high spatial variation. For that reason, additional roughness is introduced: For each surface point a stochastic realization of the density along the vertical is generated, and the sediment depth at which the density has its maximum value will constitute the new height field value. The matrix of the full integral equation is reduced to a band matrix as the interaction between the point sources on the seabed is neglected from a certain range; this allows computations on long height profiles with lengths up to approximately 25 m (at 300 kHz). The equivalent roughness approach, combined with the band-matrix approach, agrees with SAX99 data at 300 kHz.     

Effect of the substrate temperature and deposition rate on the initial growth of thin lithium niobate-tantalate films deposited from a thermal plasma

Thin lithium niobate-tantalate (LiNb_0.5Ta_0.5O₃) films are studied at the initial stage of deposition from a thermal plasma. The effect of two deposition parameters (the substrate temperature and the deposition rate) on the film morphology, the film crystallinity, and the density of nuclei growing on a (0001) sapphire substrate are investigated. It is shown that the crystalline structure and roughness of a film are determined, for the most part, in the initial growth stage and therefore depend directly on both parameters. At the optimum temperatures and growth rates for obtaining good characteristics of (0006) texture, crystallinity, and surface roughness of the films, the film nuclei on the substrate have a high density and good epitaxial orientation to it. If the growth conditions are not optimum, the islands are either amorphous or have a low density on the substrate surface. The nucleation activation energy is observed to decrease as the deposition rate increases, which supports the assumption that the species that are active in film deposition are “hot” clusters forming in an oxygen-argon plasma in the immediate vicinity of the substrate.     

浅海环境噪声垂直指向性探测海底参数

浅海环境噪声垂直指向性和海底地声参数密切相关,可以利用它来反演海底参数。本文在水平分层介质噪声场模型下,讨论了海底参数对噪声垂直指向性的敏感度,并假设半无限大液态海底,采用效率较高的自适应单纯形模拟退火算法全局寻优,根据中国某海区实测噪声数据,估计了海底声速、密度及衰减,分析了反演结果的不确定性。鉴于声速和密度在不同频率的反演结果不够一致,文中提出多频海底参数联合反演方法,提高了反演海底声学参数的精度。     

Multistage morphological segmentation of bright-field and fluorescent microscopy images

This paper describes the multistage morphological segmentation method (MSMA) for microscopic cell images. The proposed method enables us to study the cell behaviour by using a sequence of two types of microscopic images: bright field images and/or fluorescent images. The proposed method is based on two types of information: the cell texture coming from the bright field images and intensity of light emission, done by fluorescent markers. The method is dedicated to the image sequences segmentation and it is based on mathematical morphology methods supported by other image processing techniques. The method allows for detecting cells in image independently from a degree of their flattening and from presenting structures which produce the texture. It makes use of some synergic information from the fluorescent light emission image as the support information. The MSMA method has been applied to images acquired during the experiments on neural stem cells as well as to artificial images. In order to validate the method, two types of errors have been considered: the error of cell area detection and the error of cell position using artificial images as the “gold standard”.     

表面粗糙度的分形特征及其对微通道内层流流动的影响

基于分形几何学,研究了表面粗糙度的分形特征.采用Weierstrass- Mandelbrot函数对多尺度自仿射的表面粗糙度进行了描述;建立了微通道内层流流动的三维模型并对表面粗糙度的影响进行了数值模拟,分析了雷诺数、相对粗糙度和分形维数对流动阻力特性的影响.研究结果表明,与常规尺度通道不同,粗糙微通道的Poiseuille数不再是常数,而是随雷诺数近似线性增加;相对粗糙度越大,流动产生的回流和分离所导致的流动压降越明显.在相同的相对粗糙度下,粗糙表面的分形维数越大,表面轮廓变化就越频繁,这也将导致流动阻 关键词： 粗糙度 层流阻力系数 微通道 分形     

Directionality of the metallic bonding in titanium carbide

The ionic-covalent bonding in TiC is studied within a CNDO parameterized cluster model. Distribution of charge density reveals a directionality of the Ti-Ti metal bond as observed in the Ti dimer. This directionality is particularly significant at the Fermi level because of hybridisation involving d_xy ,d_xy and d_yz types of states. However, in the presence of a carbon vacancy the directionality of the metal bond is not as clear with charge spilling over into the vacant lattice site. The directional nature of the Ti-Ti bond at the Fermi level remains fairly unchanged.     

Multi-focus image fusion algorithm based on compound PCNN in Surfacelet domain

In order to effectively retain details and suppress noise, a multi-focus image fusion method based on Surfacelet transform and compound PCNN is proposed. Surfacelet transform is a powerful multi-resolution analysis tool which is able to decompose the original image into a number of different frequency band sub-images, compound PCNN model is a combined model of PCNN and dual-channel PCNN which is to select the fusion coefficients from the decomposed coefficients, the Local sum-modified-Laplacian (LSML) is selected as external stimulus of compound PCNN, fusion coefficients are decided by compound PCNN. The experimental results show that the new method has a good performance, fusion image has more texture details and it is more similar to the original images, the objective evaluation indexes show that this method is superior to the traditional image fusion methods.     

Automated oral cancer identification using histopathological images: a hybrid feature extraction paradigm

Oral cancer (OC) is the sixth most common cancer in the world. In India it is the most common malignant neoplasm. Histopathological images have widely been used in the differential diagnosis of normal, oral precancerous (oral sub-mucous fibrosis (OSF)) and cancer lesions. However, this technique is limited by subjective interpretations and less accurate diagnosis. The objective of this work is to improve the classification accuracy based on textural features in the development of a computer assisted screening of OSF. The approach introduced here is to grade the histopathological tissue sections into normal, OSF without Dysplasia (OSFWD) and OSF with Dysplasia (OSFD), which would help the oral onco-pathologists to screen the subjects rapidly. The biopsy sections are stained with H&E. The optical density of the pixels in the light microscopic images is recorded and represented as matrix quantized as integers from 0 to 255 for each fundamental color (Red, Green, Blue), resulting in a M×N×3 matrix of integers. Depending on either normal or OSF condition, the image has various granular structures which are self similar patterns at different scales termed "texture". We have extracted these textural changes using Higher Order Spectra (HOS), Local Binary Pattern (LBP), and Laws Texture Energy (LTE) from the histopathological images (normal, OSFWD and OSFD). These feature vectors were fed to five different classifiers: Decision Tree (DT), Sugeno Fuzzy, Gaussian Mixture Model (GMM), K-Nearest Neighbor (K-NN), Radial Basis Probabilistic Neural Network (RBPNN) to select the best classifier. Our results show that combination of texture and HOS features coupled with Fuzzy classifier resulted in 95.7% accuracy, sensitivity and specificity of 94.5% and 98.8% respectively. Finally, we have proposed a novel integrated index called Oral Malignancy Index (OMI) using the HOS, LBP, LTE features, to diagnose benign or malignant tissues using just one number. We hope that this OMI can help the clinicians in making a faster and more objective detection of benign/malignant oral lesions.