遗传规划多类图像分类算法研究

基于树结构的遗传规划图像分类器,仅能从树根节点输出单一实数值,要完成多类图像分类任务比较困难。提出遗传规划多类图像分类算法,等差权值中心动态边界确定分类算法和权值快速下降中心动态边界确定分类算法。在遗传规划分类器进化过程中,动态确定训练样本各类的边界,通过不同的权重设置策略,使类的边界值设定更趋合理,以提高分类器分类的准确度。图形分类测试表明,这2种算法,用于难度大的多类图像分类问题,能提高分类的准确度,分类效果理想。     

山西雁门关生态畜牧经济区良种肉用种羊舍的设计研究

结合山西省右玉县当地的地质资料、气象资料以及其它的自然条件,对山西右玉种羊场项目的种羊舍进行设计,包括总平面图、母羊舍、公羊舍、科研用房的设计。对每个项目的设计,包括确定设计参数、方案及绘制平面图、剖面图等,另外还对母羊舍的斜梁进行了受力安全验证。尤其在母羊舍设计时,引用了民用复式住宅的原理进行设计,提高了建筑的平面系数,解决母羊多时采食宽度不足的问题。该设计研究,提供了较合理、实用的优化方案。目前依据该设计方案建成的种羊舍,已通过了山西省财政厅组织的专家组的验收,并投入科研使用,使用效果较好。     

分类规则挖掘算法综述

分类规则挖掘是数据挖掘中一个重要的研究领域.通过介绍当前数据挖掘中具有代表性的分类算法,总结了各种算法的优缺点,给出了分类算法的应用以及分类算法面临的挑战,并对分类算法的发展方向进行了展望,为使用者选择算法或研究者改进算法提供了借鉴.     

优质细毛羊现行育种方案优化研究

应用动物育种规划原理和方法,对优质细毛羊现行育种方案育种效果进行优化分析,提出最优化育种方案。研究认为,当育种群基础母羊数量达到900只左右,育种群规模比例达到30%,测验群比例为25%时,每年选留15只测验公羊,其综合育种进展和育种效益最高;育种群公羊使用1年,母羊使用4年,可获得理想的育种效益。通过优化分析,群体达到最佳状态时综合育种进展可达到10.168元,育种效益为102.857元,育种投入产出比为1:7.0435。     

3个新疆绵羊品种中BMPR-IB基因的RFLP分析及比较

通过比较研究哈萨克羊、吐鲁番黑羊和策勒黑羊3个新疆绵羊群体中BMPR-IB基因的多态性,为后期分子育种提供理论依据。利用PCR-RFLP技术检测BMPR-IB基因在3个新疆绵羊群体中的多态性,并利用生物信息学软件分析其与绵羊产羔数的关联性,筛选与新疆绵羊高繁殖力性状相关的分子标记。结果显示:吐鲁番黑羊只存在1种基因型,没有多态性。哈萨克羊和策勒黑羊群体中的BMPR-IB基因具有多态,存在3种基因型:++、B+和BB,其中BB基因型个体的产羔数显著高于++基因型个体产羔数(P0.05)。所得结论为:BMPR-IB基因可能是影响哈萨克羊和策勒黑羊繁殖性状的主效基因,BMPR-IB基因是否是影响吐鲁番黑羊繁殖性状的主效基因,有待深入研究。     

AdaBoost classifiers for pecan defect classification

One of the constraints in the adoption of machine vision inspection systems for food products is low classification accuracy. This study attempts to improve pecan defect classification accuracy by using machine learning classifiers: AdaBoost and support vector machine (SVM). X-ray images of good and defective pecans, 100 each, were segmented and features were extracted. Twenty classification runs were made to adjust parameters and 300 classification runs to compare classifiers. The Real AdaBoost classifier gave average classification accuracy of 92.2% for the Reverse water flow segmentation method and 92.3% for the Twice Otsu segmentation method. The Linear SVM classifier gave average classification accuracy of 90.1% for the Reverse water flow method and 92.7% for the Twice Otsu method. Computational time for the classifiers varied by two orders of magnitude: Bayesian (10⁻⁴ s), SVM (10⁻⁵ s), and AdaBoost (10⁻⁶ s). AdaBoost classifiers improved classification accuracy by 7% when Bayesian accuracy was poor (less than 89%). The AdaBoost classifiers also adapted well to data variability and segmentation methods. A minimalist AdaBoost classifier, more suitable for real time applications, using fewer features can be built. Overall, the selected AdaBoost classifiers improved classification accuracy, reduced classification time, and performed consistently better for pecan defect classification.     

Comparison of machine learning algorithms for mapping mango plantations based on Gaofen-1 imagery

Mango is a commercial crop on Hainan Island, China, that is cultivated to develop the tropical rural economy. The development of accurate and up-to-date maps of the spatial distribution of mango plantations is necessary for agricultural monitoring and decision management by the local government. Pixel-based and object-oriented image analysis methods for mapping mango plantations were compared using two machine learning algorithms (support vector machine (SVM) and Random Forest (RF)) based on Chinese high-resolution Gaofen-1 (GF-1) imagery in parts of Hainan Island. To assess the importance of different features on classification accuracy, a combined layer of four original bands, 32 gray-level co-occurrence (GLCM) texture indices, and 10 vegetation indices were used as input features. Then five different sets of variables (5, 10, 20, and 30 input variables and all 46 variables) were classified with the two machine learning algorithms at object-based level. Results of the feature optimization suggested that homogeneity and variance were very important variables for distinguishing mango plantations patches. The object-based classifiers could significantly improve overall accuracy between 2–7% when compared to pixel-based classifiers. When there were 5 and 10 input variables, SVM showed higher classification accuracy than RF, and when the input variables exceeded 20, RF showed better performances. After the accuracy achieved saturation points, there were only slightly classification accuracy improvements along with the numbers of feature increases for both of SVM and RF classifiers. The results indicated that GF-1 imagery can be successfully applied to mango plantation mapping in tropical regions, which would provide a useful framework for accurate tropical agriculture land management.     

6个绵羊品种（系）不同血液蛋白基因型母羊间多羔性状差异性的研究

采用非连续性聚丙烯酰胺凝胶电泳检测了34只小尾寒羊、37只无角陶赛特、35只萨福克和中国美利奴（新疆型）多胎品系（29只）、肉用品系（30只）、体大品系（29只）6个品种（系）共194只母羊的血红蛋白（Hb）、血清酯酶（ES）和血清转铁蛋白（Tf）3个血液蛋白（酶）位点的多态性,并对不同基因型母羊间的多羔性状进行差异分析。结果表明,萨福克HbBB基因型产羔数高于HbAB基因型（P〈0．01）,分别为1．81±0．40和1．00±0．00只;无角陶赛特HbAA、HbAB基因型均产双羔,因而HbA等位基因与双羔性状有关;而在其余4个品种（系）不同Hb基因型母羊间的多羔性无显著差异。小尾寒羊TfBB基因型平均产羔数最高,达到3．5只;中国美利奴肉用TfAC基因型和中国美利奴体大TfAA、TfCC基因型具有多羔的趋势;而在其余3个品种（系）Tf基因型母羊间的多羔性无显著差异。中国美利奴多胎和小尾寒羊的ESa表型均具有多羔的趋势;而在其他4个品种（系）ES表型母羊间的多羔性无显著差异间。本研究表明,绵羊血液蛋白位点的多羔基因型存在着品种（系）差异性,应根据不同品种（系）各自的多羔基因型进行绵羊多羔性状的选育。     

中草药治疗母羊不孕症试验

母羊不孕症直接影响羊的繁殖能力,给养羊业带来严重的经济损失,是造成养羊业经济效益降低的重要疾病之一。应用西药治疗母羊不孕症易产生抗药、残留、畸变等毒副作用,而用中兽医理论辩证实施的中药治疗,因病施治,效果良好。为此,对试验基地的母羊不孕症临床特征进行分类,并结合中兽医气血精液的辩证理论,采用活血化瘀、清热利湿、温经散寒、补肾助阳、软坚散结等为主的组方原则进行中草药治疗,取得了较好的效果,不孕母羊受胎率显著提高。     

肉用繁殖母羊养殖生产成本分析

通过对山西省一些养殖场和养羊农户所饲养的肉用繁殖母羊养殖生产成本和经济效益的分析,估算出在现有市场行情下,山西省种羊场繁殖1只母羊的经济效益大约为907.7元,农户舍饲养殖1只繁殖母羊的经济效益约为240.1元,这为养殖场和养殖农户提供了一些养殖效益方面的数据,以便更好地推动山西省养羊业的发展。     

中亚以东南绵羊群体钾座位遗传分化的研究

利用中心产区典型群简单随机抽样法,以湖羊、同羊、滩羊、小尾寒羊、洼地羊为研究对象,引用我国周边国家、地区绵羊品种资料为分析背景,采用Na/K/Cl离子分析仪对红细胞中钾的含量进行测定。结果表明:1)各群体钾座位中立性检测的观察值除Bhyanglung绵羊外,其余均在L95和U95之间,表明钾座位作为中立性座位,基本未受到选择等因素的影响;2)血钾基因频率在地理分布上存在一定规律,基本符合将中亚以东南固有绵羊系统划分为“蒙古羊”、“藏羊”和“南亚-东南亚羊”三大集团;3)中亚以东南16个绵羊群体的钾座位基因分化系数为0.185 9。这说明上述16个绵羊群体钾座位81.41%的变异是由群体内的遗传多态现象引起的,只有18.59%的变异来自于群体间的差异。     

凉山半细毛羊种质特性的研究

对凉山半细毛羊的产毛、产肉性能、繁殖性能、生长发育、生理生化性状、血液蛋白多态性和染色体核型作了全面的介绍。凉山半细毛羊成年公羊体重 85 2 5± 3 2 5kg、母羊 4 8 11± 4 13kg ;剪毛量分别为 6 5 5± 0 12kg和4 12± 0 32kg ;毛长分别为 18 12± 0 5 1cm和 15 74± 0 2 1cm ;羊毛主体细度 4 8～ 5 0支。 6月龄肥羔胴体重16 83± 0 76kg ,屠宰率为 5 0 79%。母羊初配年龄为 10～ 12月龄 ,平均产羔率为 111 36 %。染色体数目 2n =5 4。在长期的精心选育和当地生态条件的影响下 ,凉山半细毛羊的生理生化性状产生了不同程度的变化 ,以适应凉山地区的生态条件     

Genetic differentiation and gene flow among six sheep breeds of Mongolian group in China

The level of genetic differentiation, gene flow and the relationship between geographical distance and genetic differentiation among six sheep populations of Mongolian group in China (Tong sheep, small-tailed Han sheep, Hu sheep, Tan sheep, Ujumuqin sheep and Bayinbuluk sheep) were analyzed using seven microsatellites. The trees were constructed from diversity coefficient (DC) distances among the six sheep populations. The overall heterozygote deficit across all the populations (F _it) was between 0.167 (OarAE101) and 0.044 (MAF33). The over-all significant deficit of heterozygote, because of inbreeding within breeds, (F _is) was between 0.089 (OarFCB304) and 0.005 (MAF33). The coefficient of genetic differentiation (F _st) was between 0.100 (OarAE101) and 0.022 (OarFCB48). It indicated that 3.9% of the total genetic variation could be explained by breed differences and the remaining 96.1% by differences among individuals for each population. This illustrated that most variations existed within breeds and genetic differentiation level were very low among sheep breeds of the Mongolian Group in China. The average number of effective migrants exchanged per generation (Nem) ranged from 2.7369 (Tan sheep and Bayinbuluk sheep) to 44.3928 (Tong sheep and Hu sheep), and the mean value was 11.25213. Significantly positive relationships between the level of genetic differentiation and geographical distance and genetic distances were detected. It is concluded that genetic differentiation of sheep breeds of Mongolian group in China is mainly the result of natural selection (different living conditions). __________ Translated from Journal of Yangzhou University (Agricultural and Life Science Edition), 2007, 28(3): 22–26 [译自: 扬州大学学报(农业与生命科学版)]     

宁夏农垦万只商品肉羊生产场羊群结构的优化与设计

为了稳步推进宁夏农垦企业集团百万肉羊产业化生产,笔者结合宁夏农垦生产实际,在既充分考虑宁夏农垦百万肉羊良种繁育和杂交利用体系的总体建设方案和现有建设实际,又紧密结合商品肉羊生产场目前技术力量和实际生产水平的前提下,提出了宁夏农垦万只商品肉羊生产场羊群结构优化与设计的具体方案:优秀种公羊50只,占羊只总数的0.5%,其中6月龄、1岁、2岁、3岁、4岁、5岁公羊分别占公羊总数的24%、18%、16%、16%、14%、12%;二元杂交母羊9950只,占羊只总数的99.5%,其中6月龄、1岁、2岁、3岁、4岁、5岁、6岁母羊分别占母羊总数的18.3%、16.4%、15.6%、14.5%、13.2%、11.8%、10.2%。     

敖汉细毛羊某些主要数量性状相关性的研究

本文对561只敖汉细毛羊育成母羊的资料进行了相关分析.以早期性状为自变量,以经济性状为依变量,通过逐步回归分析,得出了预测育成母羊毛长、毛量的最优回归方程式.通过对各性状间相关系数的剖分,揭示了性状间的真实关系,并绘制了估测育成母羊毛量的列线图.     

山西细毛羊品种的培育

山西毛肉兼用细毛羊是用当地的粗毛母羊和由苏联引进的高加索细毛公羊杂交培育而成的。经历了三十年的育种时间(1953——1983)。育种基地是在山西种羊场及寿阳、襄垣两县。现有符合育种指标的细毛羊111,300只。自1975年以来向省内外推广出2万余只。无论纯种繁殖或杂交改良其它的粗毛羊效果良好。本品种抗病力强,耐粗饲,能游走,善爬山,很能适应山西的自然环境。体形外貌:结构匀称,体质结实,骨骼强壮;头中等大小;母羊无角,公羊有螺旋形角或无角;母羊体躯平展,有的颈下有纵皱褶,公羊颈下有纵褶褶,或有1——2个横褶皱;体躯深、宽而长,四肢端正,被毛闭合性好,绒毛着生到两眼连线,前肢达到腕关节,后肢达飞节。羊毛品质:密度中等,细度60——64支,成年羊毛长在7厘米以上,周岁羊在8厘米以上,弯曲正常,油汗含量适中,被毛匀度良好。主要生产性能如下:     

中国蒙系6个绵羊品种的遗传分化和基因流

利用7对微卫星引物对中国蒙系6个绵羊品种(同羊、小尾寒羊、湖羊、滩羊、乌珠穆沁羊和巴音布鲁克羊)的遗传分化、基因流、遗传分化程度与地理距离间的关系进行分析,并利用遗传距离构建系统树。结果表明:6个绵羊群体中,总群体近交系数Fit最高的位点为OarAE101(0.1670),最低的为MAF33(0.0440);群体间分化系数Fst最高的位点为OarAE101(0.1000),最低的为OarFCB48(0.0220);群体内近交系数Fis最高的位点为OarFCB304(0.0890),最低的为MAF33(0.0050)。Fst平均为3.9%,即由各群体内个体间的差异引起的遗传变异是96.1%,说明遗传变异绝大部分存在于品种内,品种间的遗传分化水平很低。6个绵羊群体每世代两群体间有效迁移个体数为2.7369(滩羊和巴音布鲁克羊)~44.3928(同羊和湖羊),平均为11.2521,均反映出品种间的基因流通畅。品种间的遗传差异与地理距离呈显著相关。初步推断我国蒙系绵羊品种间的遗传分化主要是自然选择(生境异质性)作用的结果。     

A localized negative genetic correlation constrains microevolution of coat color in wild sheep

The evolutionary changes that occur over a small number of generations in natural populations often run counter to what is expected on the basis of the heritability of traits and the selective forces acting upon them. In Soay sheep, dark coat color is associated with large size, which is heritable and positively correlated with fitness, yet the frequency of dark sheep has decreased. This unexpected microevolutionary trend is explained by genetic linkage between the causal mutation underlying the color polymorphism and quantitative trait loci with antagonistic effects on size and fitness. As a consequence, homozygous dark sheep are large, but have reduced fitness relative to phenotypically indistinguishable dark heterozygotes and light sheep. This result demonstrates the importance of understanding the genetic basis of fitness variation when making predictions about the microevolutionary consequences of selection.     

鲁西黄牛肉用品系核心群育种方案遗传和经济评估

针对鲁西黄牛肉用品系现行的核心群育种方案,以育种最大投资回报率为评估标准,通过活体质量、繁殖、日增质量和胴体性状测定记录作为选择信息,根据使用不同可利用记录作为不同选种的标准设定了4种选择方案(方案1:活体质量信息+日增质量信息;方案2:活体质量信息+繁殖信息;方案3:方案1+方案2;方案4:方案3+胴体性状信息),并采用确定性模型评估了4种选择方案的遗传进展和育种效益.结果表明,4种选择方案的年综合育种进展分别为21.87、34.08、38.09和54.86元;育种效益为147.39、204.44、241.83和309.05元;投入产出比为1∶3.78、1∶4.60、1∶5.14和1∶5.77;平均世代间隔均为5.53年.以上数据表明,随着可利用信息在选择标准中的变化,遗传进展和育种效益也随之变化,选择方案4可获得较好的育种效果,所以,在现行育种方案中,采用方案4可获得理想的育种效果.     

畜禽遗传评定的微机实现方法研究

以细毛羊育种为例，阐述了畜禽遗传评定的基本原理，并结合育种信息管理系统，研究了如何利用管理系统的信息资源，进行遗传评定输出。分析了畜禽遗传评定微机实现的数据流程。并使计算育种值和多性状综合选择指数中各种性状的选择根据具体需要由育种工作者自己设定，且通过报表打印出育种值和多性状综合选择指数。整个设计由系统的主控菜单集合，最后打包到信息系统软件包中，形成用于销售的商用软件。