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.     

A multigrid accelerated hybrid unstructured mesh method for 3D compressible turbulent flow

 A cell vertex finite volume method for the solution of steady compressible turbulent flow problems on unstructured hybrid meshes of tetrahedra, prisms, pyramids and hexahedra is described. These hybrid meshes are constructed by firstly discretising the computational domain using tetrahedral elements and then by merging certain tetrahedra. A one equation turbulence model is employed and the solution of the steady flow equations is obtained by explicit relaxation. The solution process is accelerated by the addition of a multigrid method, in which the coarse meshes are generated by agglomeration, and by parallelisation. The approach is shown to be effective for the simulation of a number of 3D flows of current practical interest. Sponsored by The Research Council of Norway, project number 125676/410 Dedicated to the memory of Prof. Mike Crisfield, a respected colleague     

地震平衡技术及其在地震反演中的应用

地震平衡技术是一种在地震反演中改善井震关系的新技术。它通过单井对比、空间分形建模以及平衡处理的手段 ,克服地震反演中子波空变的影响 ,使得经过地震平衡技术处理的地震剖面井震关系得到较好改善 ,从而可以用统一的子波对研究区的地震资料进行反演     

基于扩展分形和CFAR特征融合的SAR图像目标识别

研究了多信息融合技术在SAR图像目标识别中的应用。将扩展分形特征(Extended Fractal)与双参数恒虚警特征(Double Parameter CFAR)形成的多信息进行融合处理。运用Dempster-Shafer证据理论，在决策层对SAR图像中的像素进行识别分类。实验结果表明通过融合对像素分类的准确性明显好于单特征的检测结果，减少了虚警概率，提高了系统的识别能力。     

基于体全息技术的WDM器件

利用体全息技术制作波分复用器件是一种新方法。本文介绍了此种方法的基本原理和器件的基本性能，并报道了国外对该项技术的研究进展。     

黄河黑山峡河段大柳树水库比小观音水库原始库容多40亿m3的作用

一级开发大柳树高坝原始库容比二级开发小观音高坝多 4 0亿m3 ,库容大带来的好处体现在 5～ 7月可多向下游增供水量 ,缓解下游河道断流 ;延长泄放清水时间 ;增加发电效益 ;满足宁蒙河段防凌要求 ;增加汛期集中泄水 ,减缓下游河道淤积萎缩 ;满足南水北调西线工程水量调节要求等。为了更好优化配置黄河水资源 ,从全局和长远利益考虑 ,应采用一级开发方案 ,修建大柳树高坝大库     

燃烧—气体容量法测定冶金石灰中的二氧化碳

冶金石灰是石灰石经过工业煅烧的产品。其二氧化碳含量低，且经过高温煅烧后基本不含有机物。无需按照《石灰石、白云石化学分析国家标准中二氧化碳的化学分析方法》中用酸进行试样分解，只需在1200℃高温中灼烧，使石灰中残留的或煅烧后吸收空气中二氧化碳而形成的碳酸盐分解，象通常测定含碳耐火材料中二氧化碳那样，采用气体容量法进行测定。     

Influence of ethyl methacrylate content on the volume‐phase transition of temperature‐sensitive poly[(N‐isopropylacrylamide)‐co‐(ethyl methacrylate)] microgels

A novel series of temperature‐sensitive poly[(N‐isopropylacrylamide)‐co‐(ethyl methacrylate)] (p(NIPAM‐co‐EMA)) microgels was prepared by the surfactant‐free radical polymerization of N‐isopropylacrylamide (NIPAM) with ethyl methacrylate (EMA). The shape, size dispersity and volume‐phase transition behavior of the microgels were investigated by transmission electron microscopy (TEM), ultraviolet–visible (UV–Vis) spectroscopy, dynamic light scattering (DLS) and differential scanning calorimetry (DSC). The transmission electron micrographs and DLS results showed that microgels with narrow distributions were prepared. It was shown from UV–Vis, DLS and DSC measurements that the volume‐phase transition temperature (VPTT) of the p(NIPAM‐co‐EMA) microgels decreased with increasing incorporation of EMA, but the temperature‐sensitivity was impaired when more EMA was incorporated, causing the volume‐phase transition of the microgels to become more continuous. It is noteworthy that incorporation of moderate amounts of EMA could not only lower the VPTT but also enhance the temperature‐sensitivity of the microgels. The reason for this phenomenon could be attributed to changes in the complicated interactions between the various molecules. Copyright © 2004 Society of Chemical Industry     

在UNIX工作站上实现SYBASE数据库的空间扩展

七号信令网管系统的数据库服务器采用惠普UNIX工作站及SYBASE数据库。由于采集的数据量增大，原先设计的数据库空间不能满足数据存储三个月的要求．因此采用一块外置硬盘对数据库空间进行扩展。详细地介绍了SYBASE数据库空间扩展的全过程。     

Elimination of organic matter present in wastewaters from the cork industry by membrane filtration

BACKGROUND: The first stage of the cork industrial process generates great volumes of wastewater with moderate to high organic pollutant content that must be purified using different procedures, such as filtration by membranes. RESULTS: The tangential filtration of these wastewaters was studied using two different laboratory equipments. In the first one, three ultrafiltration (UF) membranes were tested, with molecular weight cut‐off (MWCO) 100 kDa and 30 kDa, and two operating modes were used: total recycling of permeate and retentate streams, and in continuous mode, without recycling both streams. In the total recycling UF experiments, the influence of the operating variables on the permeate flux was first established. The effectiveness of the different membranes was determined by evaluating the rejection coefficients for several parameters that measure the global pollutant content of the effluent. The values found for these rejection coefficients were in the following order: ellagic acid and color > absorbance at 254 nm > tannic content > COD (chemical oxygen demand). In the continuous mode experiments, the fouling mechanism for each membrane was established by fitting the experimental data to various filtration fouling models given in the literature. The operating mode in the second equipment was batch concentration, and additional experiments were carried out with an UF membrane (2 kDa), and with a NF membrane (with MWCO in the range 150–300 Da). CONCLUSIONS: The three operating modes tested provided different rejection levels of organic matter; among them, the most effective procedure tested was batch concentration mode using a NF membrane. Copyright © 2007 Society of Chemical Industry