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.     

基于承包商预期收益不变的不平衡报价模型

在招标实践中,通常采用最低价中标原则,使得投标人不得不压低其投标价格。基于承包商预期收益不变的不平衡报价模型,是在“工程量清单报价”的条件下,考虑到施工中工程量的变化及施工顺序的影响,体现了资金的时间价值,在保持承包商预期收益不变的前提下,通过调整各分项工程的单价,来降低工程报价,以增加中标机会而建立的。所建立的模型简单,易于操作,并配以具体的工程实例,通过计算机求得其最优解,可供承包商投标报价时参考。     

人工神经网络预测煤的发热量

应用人工神经网络处理部分广东地区煤质分析数据,建立网络模型,预测出煤的发热量,经实测值与预测值比较,在一定的煤种范围内预测值与实测值比较接近。     

加快塑料技术中心建设提高新产品开发水平

对中国石油化工集团公司(简称中国石化)合成树脂生产及加工应用技术的现状进行了分析，指出中国石化应加快建设塑料加工应用技术中心，为塑料产品的加工用户提供高技术水平和快捷周到的技术服务，在合成树脂新产品开发和产品结构调整中发挥积极作用，以提高合成树脂产品的市场竞争力。     

火工品压药机装配误差的研究

针对军用雷管生产线中压药机结构,分别用极值法和概率法对压药机冲子和雷管的配合精度,进行了装配尺寸链的描述和计算,并对所得的结果进行了比较。     

甲醇氧化羰化制碳酸二甲酯产物气相色谱分析

以Chromosorb 10 1为固定相 ,气相色谱法分析甲醇氧化羰化制碳酸二甲酯产物中水、甲醇、乙酸甲酯和碳酸二甲酯等组份的含量 ,考察了柱温 ,载气流量对分离度的影响 ,确定了适宜的分析条件 ,并用校正面积归一化法定量。该方法操作简便 ,快速 ,精度高 ,重现性好     

NO emission characteristics in counterflow diffusion flame of blended fuel of H2/CO2/Ar

Flame structure and NO emission characteristics in counterflow diffusion flame of blended fuel of H₂/CO₂/Ar have been numerically simulated with detailed chemistry. The combination of H₂, CO₂ and Ar as fuel is selected to clearly display the contribution of hydrocarbon products to flame structure and NO emission characteristics due to the breakdown of CO₂. A radiative heat loss term is involved to correctly describe the flame dynamics especially at low strain rates. The detailed chemistry adopts the reaction mechanism of GRI 2.11, which consists of 49 species and 279 elementary reactions. All mechanisms including thermal, NO₂, N₂O and Fenimore are taken into account to separately evaluate the effects of CO₂ addition on NO emission characteristics. The increase of added CO₂ quantity causes flame temperature to fall since at high strain rates a diluent effect is prevailing and at low strain rates the breakdown of CO₂ produces relatively populous hydrocarbon products and thus the existence of hydrocarbon products inhibits chain branching. It is also found that the contribution of NO production by N₂O and NO₂ mechanisms are negligible and that thermal mechanism is concentrated on only the reaction zone. As strain rate and CO₂ quantity increase, NO production is remarkably augmented. Copyright © 2002 John Wiley & Sons, Ltd.     

Cif2000平台下的核磁共振测井解谱方法研究

介绍了在Cif2000多井解释平台下的核磁共振解谱方法与编程实现。解谱采用加入平滑因子后在特征矩阵的奇异值分解中截去小的非零奇异值的方法，可以在低信噪比时得到稳定的弛豫谱，在油田实际应用中证明了该方法的有效性。根据该方法在Cif2000平台上编制了完整的解谱处理程序，可以直接用于油田的生产实际。     

减压渣油与FCC油浆共炭化的化学组成变化

考察了齐鲁石化胜利炼油厂催化油浆（FCC）和胜利减压渣油（VR）在490℃，0.8MPa下不同混合比的原料和不同反应时间的中间产物的HS、TS、PS、P及其1h中间产物的HS和TS的^1H-NMR。数据结果反映了体系的反应速度。掺入FCC油浆抑制了VR的反应活性，降低了体系的反应速度，增加了基质的溶解度；1h中间产物的组成结构更接近于原料及反应体系的反应性和基质的溶解度；1h中间产物的芳香度越高，取代基越少，侧链越短，基质对VR中的活性反应组分在炭化早期生成的高度缩合物质的溶解度越大。