Identification of insect-damaged wheat kernels using short-wave near-infrared hyperspectral and digital colour imaging

Healthy wheat kernels and wheat kernels damaged by the feeding of the insects: rice weevil (Sitophilus oryzae), lesser grain borer (Rhyzopertha dominica), rusty grain beetle (Cryptolestes ferrugineus), and red flour beetle (Tribolium castaneum) were scanned using a near-infrared (NIR) hyperspecrtal imaging system (700-1100 nm wavelength range) and a colour imaging system. Dimensionality of hyperspectral data was reduced and statistical and histogram features were extracted from NIR images of significant wavelengths and given as input to three statistical discriminant classifiers (linear, quadratic, and Mahalanobis) and a back propagation neural network (BPNN) classifier. A total of 230 features (colour, textural, and morphological) were extracted from the colour images and the most contributing features were selected and used as input to the statistical and BPNN classifiers. The quadratic discriminant analysis (QDA) classifier gave the highest accuracy and correctly identified 96.4% healthy and 91.0-100.0% insect-damaged wheat kernels using the top 10 features from 230 colour image features combined with hyperspectral image features.     

稻麦籽粒计数方法研究现状及展望

作物籽粒计数是作物研究中重要的工作之一。本文在总结国内外常用的籽粒计数方法及原理的基础上,对基于计算机视觉的图像分析法、碰撞声音识别法以及基于红外线的颗粒计数方法进行综合比较、分析。结果表明,图像分析法和碰撞声音法操作简单,但易受环境影响;红外计数法被广泛使用,但籽粒下落速度过快时,受光电信息处理电路限制。针对不同的研究需求,采用不同的研究方法是比较合理的选择。     

Investigation of 3D geometry of bulk wheat and pea pores using X-ray computed tomography images

Quantitative characterization of pore topology inside grain bulks is necessary to predict the air traverse time and the cooling or fumigation pattern for the design of storage management strategies. Quantification of 3D microarchitecture of the inter-granular airspace in the grain bulks can also contribute to the development of predictive models of insect movement and for designing acoustic systems for insect infestation detection. In this study, inter-connected 3D array of void spaces was characterized by geometrical quantities such as specific surface area, pore throat size and nodal pore volume. These features were obtained from a 13.1 cm × 13.1 cm × 5 cm volume of wheat and pea bulks. The grain bulks were imaged using a high resolution X-ray computed tomography system at 200 μm resolution. The spatial distributions were computed based on 3D medial axis analysis of the void space in the images using 3DMA-Rock software and a high performance Polaris computer. The other features calculated were medial axis tortuosity, throat surface area and porosity from the 3D images. Characterization of pore throat network provides reliable observation for facilitating realistic prediction of permeability and the nature of air and gas distribution inside grain bulks.     

Neural-Intelligent Water Drops algorithm to select relevant textural features for developing precision irrigation system using machine vision

Sunagoke moss Rachomitrium japonicum is a good potential for greening material. One of the primary determinants of Sunagoke moss growth is water availability. Too much or too little water can cause water stress in plants. Water stress in plants can be detected by imaging. This study is part of on-going research aimed at developing machine vision-based precision irrigation system in a closed bio-production system for cultured Sunagoke moss. The objective of this study is to propose nature-inspired feature selection techniques to find the most significant set of Textural Features (TFs) suitable for predicting water content of cultured Sunagoke moss. The proposed Feature Selection (FS) methods include Neural-Intelligent Water Drops (N-IWD), Neural-Simulated Annealing (N-SA), Neural-Genetic Algorithms (N-GAs) and Neural-Discrete Particle Swarm Optimization (N-DPSO). TFs consist of 120 features extracted from grey, RGB, HSV, HSL and L^∗a^∗b^∗ colour spaces using ten Haralick’s textural equations. Back-Propagation Neural Network (BPNN) model performance was tested successfully to describe the relationship between water content of Sunagoke moss and TFs. Red Colour Co-occurrence Matrix (CCM) TFs, L^∗ CCM TFs, grey CCM TFs, value_(HSV) CCM TFs, green CCM TFs and lightness_(HSL) CCM TFs are recommended as individual feature-subset to be used for predicting water content of Sunagoke moss using Artificial Neural Networks. However, FS methods improve the prediction performance. The results show a significant difference between model using FS and models using individual feature-subsets or without FS. Comparative analysis shows the superiority of Neural-Intelligent Water Drops (N-IWD) compared to the other FS methods, since it achieve better prediction performance. The best N-IWD’s fitness function converged with the lowest validation-set Root Mean Square Error (RMSE) of 1.07 × 10⁻² when using 36 TFs.     

小麦主要品质性状影响面包烘烤的回归分析

以47个小麦品种为材料,测定其主要面粉理化指标、面团流变学指标和面包烘烤指标,通过相关性分析确定各品质指标间的相互关系,并进一步进行逐步回归分析确定对面包烘烤起主要作用的品质性状,以期为小麦品质育种、面包小麦品种品质指标的制定及加工原料的筛选提供参考依据。     

不同药剂拌种对小麦茎基腐病的初步研究

小麦是我国的主要粮食作物之一,由于常年秸秆还田及连作,造成小麦茎基腐病病菌有加重趋势,本文选用不同的杀菌剂进行小麦拌种,通过在小麦返青期、灌浆期、成熟期的调查,表明用6%立克秀悬浮种衣剂进行拌种防治小麦茎基腐病效果最好。     

Support vector machine approach to real-time inspection of biscuits on moving conveyor belt

An intelligent system for colour inspection of biscuit products is proposed. In this system, the state-of-the-art classification techniques based on Support Vector Machines (SVM) and Wilk's λ analysis were used to classify biscuits into one of four distinct groups: under-baked, moderately baked, over-baked, and substantially over-baked. The accuracy of the system was compared with standard discriminant analysis using both direct and multi-step classifications. It was discovered that the radial basis SVM after Wilk's λ was more precise in classification compared to other classifiers. Real-time implementation was achieved by means of multi-core processor with advanced multiple-buffering and multithreading algorithms. The system resulted in correct classification rate of more than 96% for stationary and moving biscuits at 9 m/min. It was discovered that touching and non-touching biscuits did not significantly interfere with accurate assessment of baking. However, image processing of touching biscuits was considerably slower compared to non-touching biscuits, averaging at 36.3 ms and 9.0 ms, respectively. The decrease in speed was due to the complexity of the watershed-based algorithm used to segment touching biscuits. This image computing platform can potentially support the requirements of the high-volume biscuit production.     

摩洛哥南部沿岸两种沙丁鱼耳石形态识别的初步研究

根据2011年6-8月我国拖网船在摩洛哥南部沿岸采集的46尾金色小沙丁鱼(Sardinella aurita)和53尾沙丁鱼(Sardina pilchardus)的耳石样本,利用传统形态测量法和傅立叶分析法分别进行分析,对不同的耳石形态进行判别。传统形态测量法测量了6个耳石形态参数,并利用t检验比较两种沙丁鱼耳石形态差异;耳石轮廓经图像化处理后利用软件转化为椭圆傅立叶标码(EFDs),最后分别采用6个耳石形态参数和标准化后的77个傅立叶系数进行主成分分析和判别分析。结果表明,两种沙丁鱼耳石形态在全长(TL)、宽度(TW)、背宽(DW)、腹噱长(VL)中存在显著差异(P0.01),背长(DL)和翼叶长(WL)不存在差异(P0.05);主成分分析表明,传统测量法的前两个主成分累积贡献率达77.0%,第一、二主成分的最高变量分别为全长(TL/FL)和背宽(DW/FL),在散点图中显示出了很好的区分性;而傅立叶分析法中,前16个主成分累积贡献率为82.8%,前两个主成分因子仍有较大的重叠,区分性相对较差。判别分析结果认为,传统测量法中的3个参数值进入了逐步判别分析中,总判别正确率为83.76%;傅立叶分析法的77个傅立叶系数中,有9个系数进入了逐步判别分析中,总判别正确率为92.02%。总体来看,相比传统形态测量法,傅立叶方法在鱼类耳石的判别分析中更为有效。     

地上部生长性状指标在小麦耐铝性筛选中的应用

在大容积溶液培养和小盆钵土培试验中采用地上部生长性状和根系生长性状指标对小麦基因型进行了耐铝性筛选.结果表明,供试的24个基因型的耐铝性存在极显著的差异.无论是相同或不同的筛选方法,地上部耐性指数如相对株高和相对地上部干重与根系耐性指数如相对根长和相对根系干重均呈极显著的正相关.根系耐性指数的SD、CV及分布范围较大,区分不同基因型耐铝性差异的灵敏度较高.地上部耐性指数的SD、CV及分布范围较小,反映基因型间耐铝性差异的灵敏度相对较低,但地上部也可作为可靠的耐铝性筛选指标.地上部生长性状指标尤其是株高不仅测定快速、容易,而且可以无损伤地测定而不影响后续研究的进行,可用于大规模的种质筛选或育种项目中.所采用的2种筛选方法能克服传统筛选方法如小体积溶液培养、土培和田间试验的繁琐、费时和筛选效率低的缺点,具有快速简便,一次可筛选较多样本,且条件易于控制等优点,因而大大地提高了筛选的可靠性和筛选效率.     

Textural analysis of soil images to quantify and characterize the spatial variation of soil properties using a real-time soil sensor

The primary aim of this work was to predict soil moisture content and soil organic matter using soil image texture statistics. Co-occurrence method texture statistics were used to characterize Andisol soils to extend the possibility of using RGB color space in representing composite soil color. Four co-occurrence method textural features; angular second moment (ASM), contrast (CON), correlation (COR) and inverse difference moment (IDM) calculated from generalized matrix for image texture representation were used to describe soil moisture content variation under laboratory conditions. It was found that CON and COR had negative responses to moisture content (MC) and ASM had positive response to MC. The same were also observed in direct captured field soil images in terms of textural indices against MC and soil organic matter (SOM). The correlations were significant for ASM and COR in fertilizer and combined (fertilizer-manure) plots and insignificant in manure plots. To relate sub-surface image textural indices and soil properties for individual years, stepwise multiple linear regression (SMLR) and supervised feed-forward neural networks (NN) were investigated in an attempt to provide minimal prediction errors. The improvements achieved by NN with minimal prediction errors were better than SMLR in different years. It was assumed that several years of data sets with a much larger number of observations could be used to differentiate fundamental soil properties.