Food recognition using neural network classifier and multiple hypotheses image segmentation

ABSTRACT

This paper proposes the multiple-hypotheses image segmentation and feed-forward neural network classifier for food recognition to improve the performance. Initially, the food or meal image is given as input. Then, the segmentation is applied to identify the regions, where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the features of every food item are extracted by the global feature and local feature extraction. After the features are obtained, the classification is performed for each segmented region using a feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food volume and (ii) calorie and nutrition measure based on mass value. The experimental results and performance evaluation are validated. The outcome of the proposed method attains 0.947 for Macro Average Accuracy (MAA) and 0.959 for Standard Accuracy (SA), which provides better classification performance.     

细节决定采输卤的成败

通过认真学习《细节决定成败》，结合井矿盐开采的实际，分别以生产过程中的三个细节为主，讨论了在采输卤生产技术控制中的细节的重要性。     

以一种新的距离度量作为反馈的相关跟踪方法

基于相关的模板匹配方法在图像跟踪中得到了广泛的应用,但是这种方法的主要缺点是对图像噪声非常敏感,而且由这种方法得出的相关曲面比较平坦。为了有效抑制噪声和局部遮挡对图像的影响,提出用鲁棒统计中的一个值代替相关值作为距离度量。它不是计算两幅图像的对应像素差值,而是计算两幅图像中差别不大的对应像素对的个数。实验证明这样得到的曲面相对　　收稿日期:2003 08 31;　修订日期:2003 10 28作者简介:张桂林(1944 ),男,天津人,教授,博士生导师,主要研究方向为自动识别、人体生物统计特征识别、图像处理算法与系统性能评估等。　　　　　　　要尖锐得多。当上述的距离度量应用于跟踪过程中时,模板的合理更新对跟踪非常重要。基于上述的距离度量方法,提出了一种新的模板更新策略。利用两幅图像中近似像素在整修匹配模板像素中所占的比例大小决定加权系数。这种加权策略可以自适应地根据图像序列的变化选择更新相关模板,同时在一定程度上减小模板漂移的危险。     

Extracting Light from Polymer Light‐Emitting Diodes Using Stamped Bragg Gratings

In this paper, we describe a method for increasing the external efficiency of polymer light‐emitting diodes (LEDs) by coupling out waveguided light with Bragg gratings. We numerically model the waveguide modes in a typical LED structure and demonstrate how optimizing layer thicknesses and reducing waveguide absorption can enhance the grating outcoupling. The gratings were created by a soft‐lithography technique that minimizes changes to the conventional LED structure. Using one‐dimensional and two‐dimensional gratings, we were able to increase the forward‐directed emission by 47 % and 70 %, respectively, and the external quantum efficiency by 15 % and 25 %.     

应用时频分析方法辨识时变系统的模态参数

应用Gabor展开及Hilbert变换进行时变线性系统模态参数的辨识。利用非平稳信号的Gabor变换将结构响应信号表示在时频面上，并通过Gabor展开重构分离模态分量，建立每一阶模态的时变线性模型。对单模态时变线性模型应用Hilbert变换来辨识随时间变化的模态频率和阻尼。通过对刚度和阻尼慢变的两自由度系统模态参数的仿真辨识验证辨识方法的有效性。仿真结果表明：本文方法为时变线性系统的模态参数辨识提供了一条新的途径。     

小波变换在光谱特征提取方面的应用

人们在处理高光谱图像时一般要对一些典型地物进行光谱分析、特征波段的提取,以便提取出最大量的有效信息,剔除无用或冗余的信息,然后再进行分类识别.采用小波变换的分析方法,选用合适的小波进行分解,根据分解后的高频分量中包含的重要信息,利用局部相邻的正负极值点找出对应于原始光谱曲线上每个吸收带的左右边界;利用局部过零点,即可比较精确的提取出各个吸收带的中心波长.该方法比传统的光谱特征提取方法更简洁、有效,实验证明为一种比较理想的光谱特征提取方法.     

The Two-Dimensional Clifford-Fourier Transform

Recently several generalizations to higher dimension of the Fourier transform using Clifford algebra have been introduced, including the Clifford-Fourier transform by the authors, defined as an operator exponential with a Clifford algebra-valued kernel. In this paper an overview is given of all these generalizations and an in depth study of the two-dimensional Clifford-Fourier transform of the authors is presented. In this special two-dimensional case a closed form for the integral kernel may be obtained, leading to further properties, both in the L ₁ and in the L ₂ context. Furthermore, based on this Clifford-Fourier transform Clifford-Gabor filters are introduced. AMS subject classification numbers: 42B10, 30G35 Fred Brackx received a diploma degree in mathematics from Ghent University, Belgium, in 1970 and a Ph.D. degree in mathematics from the same university in 1973. Since 1984 he is professor for mathematical analysis at Ghent University and currently he is leading the Clifford Research Group. His main interests are function theory and functional analysis for functions with values in quaternion and Clifford algebras. The research covers Clifford distributions, generalized Fourier, Radon and Hilbert transforms, orthogonal polynomials and multi-dimensional wavelets. Nele De Schepper received a diploma degree in mathematics from Ghent University, Belgium, in 2001. Since then she holds an assistantship at the Department of Mathematical Analysis of Ghent University and is a member of the Clifford Research Group. Her main interests are function theory and functional analysis for functions with values in Clifford algebras. The research covers generalized Fourier transforms, orthogonal polynomials and multi-dimensional wavelets. Frank Sommen received a diploma degree in mathematics from Ghent University, Belgium, in 1978, a Ph.D. degree in mathematics from the same university in 1980, and a habilitation degree in mathematical analysis in 1984. From 1978 until 1999 he was at the National Fund for Scientific Research (Flanders). Since 2000 he holds a Research professorship at Ghent University. His main interests are function theory and functional analysis for functions with values in quaternion and Clifford algebras. The research covers Clifford distributions, generalized Fourier, Radon and Hilbert transforms, orthogonal polynomials and multi-dimensional wavelets, algebraic analysis, hyperfunctions and radial algebra.     

电力系统谐波检测的现状与发展

准确、实时地对电力系统谐波进行检测有着重要的意义。本文根据电力系统谐波测量的基本方法，对近年来电力系统谐波检测的新方法进行了分析和评述。最后对电力系统的谐波测量进行了总结并提出了看法。     

An intelligent web-page classifier with fair feature-subset selection

The explosion of on-line information has given rise to many manually constructed topic hierarchies (such as Yahoo!!). But with the current growth rate in the amount of information, manual classification in topic hierarchies results in an immense information bottleneck. Therefore, developing an automatic classifier is an urgent need. However, classifiers suffer from enormous dimensionality, since the dimensionality is determined by the number of distinct keywords in a document corpus. More seriously, most classifiers are either working slowly or they are constructed subjectively without any learning ability. In this paper, we address these problems with a fair feature-subset selection (FFSS) algorithm and an adaptive fuzzy learning network (AFLN) for classification. The FFSS algorithm is used to reduce the enormous dimensionality. It not only gives fair treatment to each category but also has ability to identify useful features, including both positive and negative features. On the other hand, the AFLN provides extremely fast learning ability to model the uncertain behavior for classification so as to correct the fuzzy matrix automatically. Experimental results show that both FFSS algorithm and the AFLN lead to a significant improvement in document classification, compared to alternative approaches.