基于直方图局部信息和Gath-Geva模糊聚类的彩色图像分割方法

针对Gath-Geva模糊聚类算法对初始给定的聚类中心等先验信息较敏感,提出了一种基于直方图局部信息的模糊Gath—Geva聚类新算法．实验结果表明,新算法在彩色图像分割方面,与传统模糊C—Means算法相比,具有较强的分割精度．     

基于机器视觉的菊花生长自动无损监测技术

本研究采用机器视觉及图像处理技术对多株植物生长信息进行了监测研究。该监测系统使用的CCD相机被安装在四周均布10株被测植物的精密旋转云台上,相机的镜头前安装了近红外滤片,在相机顶部安装了近红外光源。旋转云台按一定时间间隔自动定角度旋转,旋转到设定角度后,CCD相机对被测植株进行取像。利用该系统,可以长时间自动连续对多株植物进行监测。为了提高测量精度,实验过程中在被检测对象的茎节叶腋处粘贴反射亮片,在图像分析处理过程中,采用一种基于亚像素和区域匹配的误差消除估计算法EEC图像分析的处理方法,有效地提高了检测精度。并且,通过研究发现在图像匹配过程中,参考模板的最小尺寸应大于15×15 pixel。利用该系统对菊花的茎长生长进行了监测分析,发现白天菊花的茎长生长率要小于夜间的生长率。该研究为利用机器视觉技术连续非接触监测植物的生长信息提供了技术理论基础。     

高粱单粒播种技术要求及与2BJ-5精播机匹配试验

为减轻高粱种植后需间苗的繁琐劳动,山西省农业科学院高粱研究所与经纬机械(集团)通用机械公司合作研制出了2BJ-5精播机。为实现该机的高粱单粒点播,根据高粱单粒点播技术要求,在大田对该机进行了匹配性试验。结果表明,2BJ-5精播机播种高粱完全可以实现单粒点播,以播深3～5 cm、株距11～13 cm、行距50 cm为宜,种子发芽率85%以上。     

大跨钢管混凝土拱桥施工监控研究

以衡南湘江公路大桥为例,探讨了钢管混凝土拱桥施工监控的主要工作内容和监控方法.指出监控系统的建立和理论分析中需要注意的问题,以此指导该桥的施工监控.最后给出了该桥的施工监控成果,并对该类桥型的施工提出一些有益的建议.     

The effect of food competition and relative size of group members on growth of juvenile gilthead sea bream, Sparus aurata

The effects of food competition and relative size of group members on growth of sea bream, Sparus aurata were tested in groups composed of four fish. Each group consisted of a medium sized test fish (approximately 250 mg) and three fish larger or smaller than the test fish. In the first experiment food competition was addressed directly by contrasting its presence with its absence. Neither food competition nor relative size of school members had a significant effect on growth. However, there was a highly significant interaction between these two main effects. Under conditions of food competition the medium sized test fish had a three times higher specific growth rate when surrounded by small vs large fish. Under conditions of no food competition when the test fish were separated during feeding a similar but reverse pattern occurred due to differences in food conversion efficiency. In the second experiment, the effects of food quantity and relative size of group members were tested. Both main effects and their interaction were highly significant. Under conditions of low food competition (i.e., surplus food supply) fish grew again better when surrounded by larger fish as compared to their growth rate when surrounded by smaller ones. Size matching via a differential growth rate in a schooling species such as sea bream was suggested to increase the confusion effect by reducing the ability of a predator to focus on an odd sized fish. The implications of direct competition for food for the efficient culture of this species are discussed.     

蓝靛果忍冬扦插繁殖技术研究

采用ABT生根粉质量浓度、处理时间、扦插基质三因素三水平正交试验设计进行蓝靛果忍冬扦插试验,结果表明:ABT生根粉质量浓度和扦插基质对生根率有显著影响,处理时间影响不显著;适宜的组合为ABT200 mg·L-1+处理时间4 h+基质为1/5珍珠岩+4/5河沙,平均扦插生根率达到86%以上。在圃地对扦插苗进行移栽,成活率达到90%以上。     

海南水稻机插泥浆育秧配套技术

海南省水稻机插秧育秧所用营养土,以红砖土或黄;沙土混合为主,并添加各种壮秧剂配制而成;不仅 育秧成本高;而且面临营养土供应及营养土运费问题.以泥浆渊水稻土冤为营养土,参考国内外泥浆育秧技术袁以省 工\省力.节本为基础结合海南特色气候条件,通过多次试验,建立一套水稻机插育秧配套技术.     

基于点匹配的区域拷贝篡改检测

图像区域拷贝是一种常见的数字图像篡改技术,目前的大部分数字图像区域拷贝取证技术未考虑旋转和缩放因素。提出一种新的基于点匹配的图像区域篡改检测算法。首先利用尺度不变旋转变换（SIFT）寻找图像中的关键点,使用主成分分析法(PCA)对关键点进行降维描述,然后利用关键点特征向量的相似度寻找关键相似点。实验表明,该算法不但能够较精确地定位出复制和粘贴的图像伪造区域,还能有效抵抗噪声污染、有损压缩以及旋转等攻击,并有效地减少运算量,提高了检测效率。     

The k-nearest neighbor technique with local linear regression

In a standard k-nearest neighbor (kNN) technique, imputations of unit-level values in the variables of interest (Y) are based on the k-nearest neighbors in a set of reference units. Nearest is defined with respect to a distance metric in the space of auxiliary variables (X). This study evaluates kNN imputations of Y with a selection, by the same distance metric, of k-nearest locally weighted regression models. Imputations are obtained as predictions using the X values of the k-nearest neighbors in the population. In simulated random sampling from three artificial multivariate populations and two actual univariate populations and sampling units composed of a single population element or a cluster of four elements, the new kNN technique: (1) improved the correlation between an imputation and its actual value; (2) lowered the root mean square error (RMSE) of imputations; (3) increased the slope in regressions of actual y values regressed against their imputed values; (4) performed relatively best with k values of 4 and sample sizes of 200 or greater; (5) compared favorably with a recently proposed kNN calibration procedure; and (6) had a higher (15–28%) RMSE than with a simple local linear regression. Distribution matching had a consistent negative effect (+10%) on RMSE.     

基于压缩感知SAMP算法的改进

压缩感知是近年来兴起的以较大压缩比率并且能够以较好的效果恢复原始信号的一种新型理论。本文针对SAMP算法在重构精度相对较差以及运算时间较长的问题,提出一种新的SAMP改进算法。该算法首先将稀疏信号进行分块后再重构,其次在重构算法过程中开始对信号稀疏度进行初始估计,然后采取变步长增加稀疏度的方法逐步求出原始信号的最优估计值。理论分析和实验表明,改进后的算法减少运算时间,提高重构精度,增进实际使用价值。