酿酒原粮大米中碎米含量测定的不确定度评定

利用大米外观测定仪,通过图像处理技术得到米粒的图像并判断米粒的大小、整米、大碎米和小碎米,使用大米外观测定仪来自动计算大米碎米含量的实验。列举出实验中可能影响检测结果各个分量的不确定度,建立合成不确定度的数学模型,计算和评估各个分量的不确定度对检测结果的影响。从实验的计算数据分析来看,影响测定酿酒原粮大米中碎米含量的不确定度其最大分量来源于样品重复性测定和仪器测定的精密度,因此选择精度高的大米外观测定仪器、增加平行测定的次数、提高操作水平是保证实验数据准确度的关键。     

基于图像识别的整精米自动检测研究

本文采取动态阈值分割，把米粒图像变成二值图像，根据区域内部像素的连通性，将不同的区域赋予不同的标记，计算出单个米粒所占的像素，再通过先期计算的整精米长度／面积比换算出米粒长度，最后根据米粒长度判断整精米和碎米。研究结果表明，该算法的自动检测与人工检测相关性大于99％，可用于整精米的自动检测。     

MCDX长度分选机

该机由扬州市正大机械制造有限公司生产，利用米粒整米和碎米长度不一致的原理，将整米和碎米分离开。与传统的白米分级筛进行组合，可以做到整米中不含碎米，碎米中不含整米。本机采用多筛筒组合结构，具有占地面积小、筛筒更换容易、操作方便等特点。     

浅谈酿酒原料大米碎米含量测定方法的适用性

采用GB/T 5503—2009《粮油检验碎米检验法》测定酿酒原料大米的碎米含量和小碎米含量,通过筛分法和人工挑选相结合的方法,分选出样品中的碎米,称量碎米质量,来计算碎米含量,也可通过图像处理技术得到米粒的图像并判断米粒的大小、整米、大碎米和小碎米,来自动计算大米碎米和大米小碎米的含量,验证其方法应用于酿酒原料大米的碎米含量和小碎米含量测定的实用性、重复性、准确性、可行性和可靠性,以及确认《粮油检验碎米检验法》在本公司实验室适用且有效。     

基于OTSU 图像分割算法的碎米检测

碎米作为大米加工过程的常见产物,常会对产品的口感、味道产生影响,因此针对整米中碎米的有效筛分尤为重要。针对上述问题,该文建立基于大津法（maximal variance between clusters,OTSU）图像分割算法的逻辑回归模型用以检测整米中的碎米。将检测结果与国标法进行对比,结果表明逻辑回归模型的曲线线下面积（area under the curve,AUC）值为0.987,柯尔莫可洛夫-斯米洛夫（Kolmogorov-Smirnov,KS）值为0.909,0.5 为最佳阈值;而国标法的AUC 值为0.922,KS 值为0.669,21 为最佳阈值。该文所建立的逻辑回归模型的准确率、精确率、召回率及F1 分数均高于国标法。此外,逻辑回归模型的AUC 值比国标法的AUC 值更接近于1,KS 值也更高,表明逻辑回归模型能够更好地区分碎米与整米。长轴（x1）、面积（x2）、短轴（x3）与长短轴比（x4）4 个特征参数都是模型中具有显著影响的因素,对应的线性关系为z=-139.97-5.35x1+10.93x2+2.86x3+34.59x4。     

MJXT滚筒精选机

该机是利用完整米和碎米长度不一致的原理，将整米和碎米分离。具有可随意控制整米含碎量、不易损伤大米表面光洁度等特点。可依据产量和选别要求，串联使用。     

小麦籽粒外观形态特征测定技术研究

利用计算机图像处理技术,通过设计图像捕捉装置获取小麦籽粒图像,编程实现小麦籽粒图像分析,能够准确测定小麦投影面积、长轴、短轴、椭圆度、等效直径等形态参数.研究结果表明:计算机图像处理系统测定小麦籽粒长轴、短轴、投影面积等参数与手工测定值均存在良好的线性关系,R2分别达到了0.991、0.993、0.985,实现了小麦籽粒外观形态参数量化测定,能够精确描述小麦形态特征.     

基于大米外观品质分析系统对碎米率的快速检测

用实验室自制开发的一种大米外观品质分析系统,对碎米进行识别测定.以国标方法为基准,调试大米外观品质分析系统测定碎米时的最佳阈值,再对实际样品进行测定.结果表明:分析系统对大米图像平均检测时间为5 s,大米碎米率和小碎米率测定的最佳阈值分别为0.68、0.40,在此阈值下测定本批大米碎米率为7％～8％,小碎米率小于0.5...     

基于双参数阈值的蚕茧分割及双宫茧识别研究

针对粘连蚕茧难以准确分割、现有双宫茧识别方法局限性等问题,本文提出了基于双参数阈值的蚕茧分割计数、蚕茧形状复原及双宫茧识别的串联算法。首先,应用可变分割阈值结合形态学运算的方法对图像进行分割计数,再遍历完成分割的每个蚕茧的像素点,应用基于边缘检测与椭圆拟合的蚕茧形状复原方法拟合蚕茧形状,然后计算蚕茧长短轴比值和面积参数,采用双参数多级阈值对双宫茧进行检测。实验结果显示,图像分割计数准确度为100%,双宫茧全部检出,双宫茧与单宫蚕茧分类准确率为98.6%。该方法检出了面积阈值法易漏检的小双宫茧,也较准确地识别出与长短轴比值阈值接近的蚕茧类型。     

机器视觉在稻米整精米率快速计算中的应用

将机器视觉技术以及图像处理方法应用于稻米整精米率计算,通过图像预处理仅保留图像中大米的轮廓信息,并采用闭合图形边缘搜索方法提取大米轮廓像素点。由于米粒形似椭圆,并以长轴为对称轴来实现米粒长度的快速计算。实验结果表明,该算法检测与人工检测相关系数为0．96,可用于整精米率的快速检测。     

An indirect approach for egg weight sorting using image processing

An indirect approach for egg weight sorting, using image processing, is proposed in the paper. The eggs are sorted in four classes by weight. Regression analysis is used for approximation of relationship between egg weight and egg geometric parameters—perimeter, area, major and minor axis, shape index and shape factor. The values of the geometric parameters, collected by image processing and the one, collected by traditional method, are compared for each egg sample, using percent differences between data. The experimental results show that the most significant parameter for egg weight indirect measurement is the egg area, with correlation coefficient 0.989. The mathematical model for the relationship between weight and area of the egg is defined with coefficient of determination 0.978. The classification accuracy is achieved within the eggs test sample sorting. The total classification error is 2.5% for test set and 12.5% for training set.     

Deformation and Dehulling of Sponge Gourd (Luffa aegyptiaca) Seeds

This work presents the compressive strength properties of sponge gourd (Luffa aegyptica) seeds to facilitate the design or adaptation of an appropriate dehuller. The sizes and shape indices of the seed and kernel, and the clearance of the kernel from the seed coat were determined. The seed samples were subjected to uni-axial quasi-static compression tests at 1.0 mm min^?1 along the minor and the major axes. The fracture resistance, stiffness modulus, modulus of elasticity, toughness, and maximum elastic deformation of the seed were obtained from the force-deformation curve. The geometric mean diameters varied from 4.0 to 4.5 mm for the seed and 1.6 to 3.9 mm for the embedded kernel; while their corresponding sphericities were 0.64 and 0.62, respectively. The compressive strength of the seed varied with loading orientation. The seed exhibited larger deformation but lower stiffness along the major axis than the minor axis. The force required for rupturing the hull were 95 N along the major axis and 81 N along the minor axis; while the corresponding energy required were 95 and 40 mJ.     

基于计算机视觉的大米整精米率检测

提出了一种在线的基于计算机视觉的大米整精米率检测的新方法。采用最大方差阈值分割对大米图像进行处理,再对分割结果进行形态学操作,实现连接着的大米的分离,最后对分离后的大米二值图像进行面积和周长特征的提取。根据米粒周长像素数目的大小采取不同识别模式,当大米周长的像素数目大于某一固定值(先验值)时,即该种大米是长粒型,采取周长识别模式;短粒型大米则采取面积识别模式。通过使用这一能够智能选择识别模式的检测方法,能够大大提高整精米率的检测效率。     

Effects of fluidized bed drying temperature and tempering time on quality of waxy rice

Waxy rice, which is soft and sticky in nature, can be used as a raw material to produce many food products. After being harvested, high-moisture waxy paddy must be dried to appropriate moisture content to prolong its storage life and to achieve higher head rice yield. Fluidized bed dryer could be used to dry waxy rice at high-temperature. However, due to the high heat and mass transfer rates during drying, stresses are generated in a rice kernel, leading to crack and low head rice yield. Tempering is thus recommended to reduce the moisture-induced stresses in the kernel after rapid drying. In this study, the effects of fluidized bed drying temperature (90, 110, 130 °C) and tempering time (30–120 min) on the quality of waxy rice, i.e., head rice yield, thermal properties, pasting properties, color, translucent kernel and microstructure, were investigated. The results showed that head rice yield of waxy rice after drying was significantly lower than that of the reference sample even when tempering was performed. Higher drying temperatures led to higher head rice yield while the tempering time did not have any effect on the head rice yield except when the drying temperature of 90 °C was used. Drying at higher temperatures also affected the starch granule morphology and the pasting properties. Waxy rice changed its appearance from opaque white to translucent when being dried at 130 °C.     

对广东省三种籼稻谷的品质分析

本文对广东省内产的3种籼稻谷（中二软占、特籼占25、珍桂矮1号）质量进行品质分析（出糙率、不完善粒、整精米率、垩白粒率、垩白度、粒型、胶稠度、直链淀粉含量、食味品质）。结果表明：3种籼稻谷质量存在明显差异。     

Effects of moisture content and compression axis on mechanical properties of Shea kernel

Some mechanical properties of Shea kernel were investigated in this study. The kernels were divided into two categories sizes namely: small size kernel (SSK) and large size kernel (LSK) and the properties investigated were: rupture force, deformation at rupture and energy consumed at rupture. The tests were carried out at a deformation rate of 50 mm/min and four moisture content levels of 25.9%, 11.60%, 6.88%, 4.98% (db) for SSK and 11.19%, 6.21%, 5.78% and 2.77% (db) for LSK. The variations in these properties were observed considering the effects of moisture content and compression axes on them as the kernels were air-dried. Sample kernels were compressed along the orthogonal axes corresponding to major axes (length), intermediate axes (width) and minor axes (thickness) of Shea kernel. Some physical characteristics of Shea kernel such as dimensions, geometric mean diameter and mass were also evaluated. Results showed that generally, rupture force, deformation and energy at rupture decreased as moisture content decreased. The regression models that best fitted the relationships were polynomial functions of the second order. The highest and lowest forces for Shea kernel to rupture were those through the minor axis (thickness) and major axis (length) respectively. These properties are often required for the design of transportation, storage and grading/sorting machines and other post harvest machines for Shea kernel.     

Effects of Ageing on Selected Cooking Quality Parameters of Rice

Ageing can improve cooking quality of rice by influencing major cooking quality parameters i.e., kernel expansion, water absorption, alkali digestion value, and gelatinization temperature along with changes in internal structure of rice grains. In this research, the effects of natural and artificial ageing on the selected cooking quality parameters of two Malaysian rice cultivars, named Mahsuri and Puteri, were studied. A relation was observed between water absorption and elongation ratio in both varieties under different aging conditions. Alkali digestion value and gelatinization temperature were also influenced by varieties and ageing conditions. This study revealed the potentiality of ageing for the improvement of rice cooking quality.     

A Deep Convolutional Neural Network Architecture for Boosting Image Discrimination Accuracy of Rice Species

In this paper, we proposed a machine vision system based on deep convolutional neural network (DCNN) architecture for improving the accuracy of classifying three distinct groups of rice kernel images compared with the traditional approaches. The main advantage of the presented method was able to avoid many heuristics as well as manual labor to tune complex parameters according to the domain to reach a modest level of accuracy in the classical feature extraction algorithms. We trained our models using stochastic gradient descent with momentum of 0.9 and weight decay of 0.0005 to optimize the network parameters and minimize the back-propagation error on the training dataset. We used a batch size between 15 and 150 and epochs time configured between 10 and 25. The experiment results showed that the highest accuracy of 99.4% obtained in the training process with batch size of 15 and epoch time of 20. We also compared the DCNN method with the traditional hand-engineered approaches of PHOG-KNN, PHOG-SVM, GIST-KNN, and GIST-SVM for rice kernel classification. The results showed that DCNN routinely outperforms other methods in similar machine vision tasks. The prediction accuracy results for test datasets by PHOG-KNN, PHOG-SVM, GIST-KNN, and GIST-SVM models were 89.1, 76.9, 90.6, and 92.1%, respectively. The highest prediction accuracy of DCNN is 95.5%, which showed the effectiveness of our proposed method for rice kernel classification. The aim of this study is to set up an automatic and accurate intelligent detection system and offering much value to current rice processing industry. With the comparably high classification accuracy, developed neural network could be used as a tool to achieve better and more objective rice quality evaluation at trading points within the rice marketing system.     

Identification of Broken Rice Kernels Using Image Analysis Techniques Combined with Velocity Representation Method

The percentage of whole kernels remaining after milling is one of the most important physical characteristics of rice quality. A method based on flatbed scanning and image analysis was developed for the identification of broken rice kernels. Velocity representation method was developed for pattern recognition based on the contour characteristics of the rice kernels. The similarity of the boundary features of the image of rice kernel was measured by similarity coefficient, which was used to identify the broken rice kernel by comparing with threshold. High recognition rates for three rice varieties were reached by this method with 96.7% for Thailand rice, 98.73% for Pearl rice, and 97.14% for Changlixiang rice, respectively, and the recognition rate could be improved by the adjustment of the similarity coefficient threshold. Because the comprehensive boundary features were the basis for the classification, this method could be more accurate compared to other methods using the single dimension feature.