Wood inspection with non-supervised clustering

Abstract. The appearance of sawn timber has huge natural variations that the human inspector easily compensates for mentally when determining the types of defects and the grade of each board. However, for automatic wood inspection systems these variations are a major source for complication. This makes it difficult to use textbook methodologies for visual inspection. These methodologies generally aim at systems that are trained in a supervised manner with samples of defects and good material, but selecting and labeling the samples is an error-prone process that limits the accuracy that can be achieved. We present a non-supervised clustering-based approach for detecting and recognizing defects in lumber boards. A key idea is to employ a self-organizing map (SOM) for discriminating between sound wood and defects. Human involvement needed for training is minimal. The approach has been tested with color images of lumber boards, and the achieved false detection and error escape rates are low. The approach also provides a self-intuitive visual user interface. Received: 16 December 2000 / Accepted: 8 December 2001 Correspondence to: O. Silvén     

Investigation of the Correlation between Texture and Microstructure on a Submicrometer Scale in the TEM

In recent years the investigation of local texture and microstructure by analysis of electron backscatter diffraction patterns (EBSP) in the SEM has become a very powerful and popular method. With the introduction of SEM with field emission guns (FEG) the spatial resolution of EBSP measurements could be enhanced from 500 nm with a tungsten emitter to better than 50 nm. This evolution of SEM techniques raises the question whether transmission electron microscopy (TEM) still has fields of application in texture research. The present article answers this question with a clear “yes” and presents three examples of investigations where TEM is indispensable. The three examples comprise the investigation of the correlation between dislocation structure and deformation texture, a study on nucleation mechanisms of recrystallization in highly deformed metals and the investigation of microtexture and microstructure in nanocrystalline materials. Together with the presentation of these cases some of the necessary measurement techniques are described briefly. It is shown that TEM has to be applied when highest spatial resolution of orientation determination and imaging and high accuracy of orientation determination are to be reached, when the three‐dimensional and quantitative characterization of lattice defects is required or when materials with a high density of lattice defects are to be investigated.     

Real-time homogenous translucent material editing

This paper presents a novel method for real-time homogenous translucent material editing under fixed illumination. We consider the complete analytic BSSRDF model proposed by Jensen et al. [ [JMLH01] ], including both multiple scattering and single scattering. Our method allows the user to adjust the analytic parameters of BSSRDF and provides high-quality, real-time rendering feedback. Inspired by recently developed Precomputed Radiance Transfer (PRT) techniques, we approximate both the multiple scattering diffuse reflectance function and the single scattering exponential attenuation function in the analytic model using basis functions, so that re-computing the outgoing radiance at each vertex as parameters change reduces to simple dot products. In addition, using a non-uniform piecewise polynomial basis, we are able to achieve smaller approximation error than using bases adopted in previous PRT-based works, such as spherical harmonics and wavelets. Using hardware acceleration, we demonstrate that our system generates images comparable to [ [JMLH01] ]at real-time frame-rates.     

A visual news processing environment

CPP-TRS (Tonfoni 1989–94) is both a methodology and a language. Through the methodology (CPP — Communicative Positioning Program) those invisible aspects of communication are identified and represented by the meta-language (TRS — Text Representation Systems), which complements natural language. CPP-TRS is a new paradigm which applies to any communicative context.Icons in CPP-TRS are not intended to represent words or sentences in order to substitute them; they are rather intended to represent visually what natural language does not convey naturally. The improvement of communication is a challenge of the present and the future: CPP-TRS is designed explicitly for it.The CPP-TRS system thus is able to support the user with a set of visual tools, that are specifically suited to structuring text and communicating effectively.The same tools can support communication very effectively in situations where ambiguity can highly compromise the final result.The system may be used in a way that is either opaque or transparent to the reader of the text.Author of CPP-TRS¢ and CPP-TRS¢ Applications Designer.     

多水平外区抑制的轮廓检测

提取自然图像中的物体轮廓是机器视觉研究的重要问题,主要困难在于自然图像中的纹理性边缘严重干扰了物体轮廓的提取。研究表明视皮层方位选择性神经元的非经典感受野机制使得人类视觉系统在处理自然图像时不仅能够抑制纹理性边缘,而且能够增强物体的轮廓。基于此人们提出多种仿生轮廓检测算法,但算法中被称为抑制水平的参量在取值较高时会漏检部分轮廓,而在其取值较低时又会引入过多的纹理性边缘。针对这一问题,提出多水平外区抑制轮廓检测算法,通过整合各级单水平外区抑制的检测信息,有效抑制了纹理性边缘和降低了漏检轮廓的可能性。实验结果表明,相对于传统算法,新算法在轮廓检测性能上提高了10%左右,并具有更好的稳健性。     

Simulation of polycrystalline aluminum tensile test with crystal plasticity finite element method

The crystal plasticity was implemented in the finite element method（FEM） software ABAQUS through the user subroutine UMAT. By means of discretizing the space at the grain level with the Voronoi diagram method, a polycrystal model was built and used in the FEM analysis. The initial orientation of each grain was generated based on the orientation distribution function（ODF）. The developed model was successfully applied in simulation of polycrystalline aluminium samples deformed by the tensile tests. The theoretical strain--stress relation was in good agreement with the experimental result. The simulation results show that the grain size has significant effect on the deformation behavior. The initial plastic deformation usually occurs at grain boundaries, and multiple slip often results in an enhanced local hardening at grain boundaries.     

取向硅钢晶粒取向的不均匀分布及其检测

利用X射线衍射和背散射电子衍射检测了取向硅钢板表层和中心部位的宏观织构和微观晶粒取向分布,观察了相应的晶粒组织。结果表明,二次再结晶不彻底的成品钢板的表面会残留非{110}<001>取向的小晶粒,并影响钢板的软磁性能。取向硅钢板表层的晶粒取向分布能够在一定程度上反映出钢板的质量,因此可以借助X射线表面无损检测技术发展取向硅钢织构的工业在线检测技术。     

一种新颖的低层视觉特征到高层语义的映射方法*

通过在低层视觉特征上提取图像的局部颜色和纹理特征,再利用模糊神经网络建立低层视觉特征和高层语义特征之间的映射,实现了图像属性信息的有效传递和可靠检索.最后,为了提高检索效率,把相关反馈引进到图像检索系统中.实验证明,该方法取得了较好的检索查全率和准确率.     

锡或锑在无取向电工钢中的研究进展

近年来,在节能减排背景之下,国内外众多研究者对无取向电工钢磁性能的提升做了大量研究。为了探索无取向电工钢磁性能提升的方法,对锡或锑对无取向电工钢磁性能的作用机制(晶粒尺寸和晶体织构的控制)进行分析。基于该作用机制,介绍锡或锑的添加对无取向电工钢磁性能的影响。经研究发现,适量的锡或锑在晶界偏聚,不会阻碍晶界的移动并且致使晶粒尺寸降低;与此同时,锡或锑在晶界偏聚不仅抑制{111}织构在原始晶界处形核及生长,还降低(100)晶粒表面能,促进(100)晶粒生长。因此,适量添加锡或锑,可使无取向电工钢铁损下降、磁感提升。最后结合生产工艺,建议无取向电工钢的研究方向应为稀土含量对高牌号无取向硅钢夹杂物尺寸和数量分布的影响,锡或锑的添加量和常化工艺参数(常化时间、常化温度)对常化晶粒尺寸的影响。