The Machine Recognition for Population Feature of Wheat Images Based on BP Neural Network

Recognition and analysis of dynamic information about population images during wheat growth periods can be taken for the base of quantitative diagnosis for wheat growth. A recognition system based on self-learning BP neural network for feature data of wheat population images, such as total green areas and leaves areas was designed in this paper. In addition, some techniques to create favorable conditions for image recognition was discussed, which were as follows: (1) The method of collecting images by a digital camera and assistant equipment under natural conditions in fields. (2) An algorithm of pixei labeling was used to segment image and extract feature. (3)A high pass filter based on Laplacian was used to strengthen image information. The results showed that the ANN system was availability for image recognition of wheat population feature.     

Apple mealiness detection using fluorescence and self-organising maps

The chlorophyll fluorescence kinetics of ‘Jonagold’ and ‘Cox’ apples, stored under different conditions to induce mealiness, were measured. Three different storage conditions were considered causing three mealiness levels: not mealy, moderately and strongly mealy. Also destructive measurements of the texture (firmness, hardness, juice content and soluble solids content) were done. Classification into different mealiness levels based on the fluorescence measurements was more performant than a classification based on the destructive measurements. To estimate the mealiness level in a non-destructive way from the fluorescence features, a number of different classifiers were constructed. Quadratic discriminants and supervised and unsupervised neural networks were tested and compared. The self-organising map gives promising results when compared with the multi-layer perceptrons and quadratic discriminant analysis. The different advantages of the constructed classifiers suggest that fluorescence can be used in an automatic sorting line to assess certain types of mealiness.     

几种白蚁种间竞争与种内相容性初步研究

将林地挖巢获得的几种白蚁,按照一定比例配置于同一培养皿中观察其相容情况,通过实验可以发现：在种间斗争中,４种白蚁争斗性强弱依次可表现为：长泰乳白蚁＞黄翅大白蚁＞囱土白蚁＞长长翅散白蚁。对于同种不同巢群的个体,黄翅大白蚁种群的 体间能够良好相容,并巢合群完全可能成为其实现种群繁衍的第三种自然演替方式。     

支持向量机在高压绝缘子污秽程度评定中的应用

高压绝缘子污秽程度的评定可以表述为多类模式识别问题,由于影响绝缘子表面污秽状况的因素复杂,并且实际获得的样本数据有限,因此传统的智能方法往往达不到工程应用的精度要求。研究利用支持向量机在解决小样本、非线性及高维识别中的优势,将可用于多类模式识别的DAG SVM s模型用于高压绝缘子污秽程度的评定。该模型易于实现,且能够找到模式间的最优分类超平面,泛化能力较高。用SVM评定高压绝缘子污秽程度时,可以综合考虑不同污秽程度绝缘子泄漏电流的电气特性、环境参数与绝缘子污秽程度之间的非线性关系,从而实现具有极大模糊性的绝缘子表面污秽程度的评定。结果表明,此方法对解决绝缘子污秽程度的评定问题具有良好的适应性和实用性。     

组培苗识别算法在基于TMS320C6711图像处理平台上的实现

为达到组培苗分割移植机器人的嵌入式控制系统对图像处理器小型化和快速化的要求 ,采用基于TMS32 0C6 711的嵌入式图像处理平台 ,开发了运行于该平台的软件系统 ,实现了组培苗识别算法在该平台上的移植。针对DSP软件开发难度大、涉及因素多 ,以及在本研究中需要大量数据存取的情况 ,在保证功能的前提下尽量精简软件结构 ,编写了轻量简洁的Bootloader和适应大量数据存取的连接命令文件。对软件进行了优化 ,将算法中占用时间较长的环节用线性汇编语言编写 ,使总处理时间降低了近 80ms。该图像处理器实验室测试结果为 ,苗节点识别准确率 96 % ,每棵苗的处理时间为 0 7～ 0 8s ,10× 2 4h连续运转稳定 ,性能达到设计要求。该图像处理器的体积只有PC机的 1/ 2 0 ,达到了小型化的要求。     

基于整枝抚育目的的立木枝干自动识别研究

为了适应我国人工工业用材林自动化整枝抚育的需要，该文提出一种用于工业用材林自动整枝的立木枝干动态识别系统框架，对人工林侧柏的枝干进行了数字图像采集及处理，利用计算机视觉、图像处理、小波分析技术对图像进行压缩、滤波、分割、消噪及边缘特征提取.同时提出了一种立木枝干计算机自动识别算法，提取立木图像枝干形状、尺寸、弯曲度及相对空间位置关系基本生长特征，即利用模式识别技术验算其与特征数据库的匹配情况，从而达到立木枝干自动识别的目的，为后续的自动识别研究提供了依据.      

G protein signalling involved in host recognition and mycoparasitismrelated chitinase expression in Trichoderma atroviride

Mycoparasitic species of Trichoderma are commercially applied as biological control agents against various fungal pathogens. The mycoparasitic interaction is host specific and includes recognition, attack and subsequent penetration and killing of the host. Investigations on the underlying events revealed that Trichoderma responds to multiple signals from the host (e.g. lectins or other ligands such as low molecular weight components released from the host's cell wall) and host attack is ac…     

黄瓜采摘机器人远近景组合闭环定位方法

针对黄瓜采摘机器人远景定位精度不高,以致切伤果实和茎蔓的问题,设计了一种基于机器视觉具有空间位置反馈功能的末端执行器。对温室环境下黄瓜果实采摘区域图像信息获取方法加以研究,综合HIS色彩空间H、S分量进行阈值分割,结合RGB色彩空间G通道边缘分布特征以及黄瓜形状特征,提取黄瓜采摘区域。基于摄像机线性透视模型,研究了采摘切割点空间定位方法,最终向采摘机械臂控制器反馈位置微调信息。采用远近景组合闭环定位方法,对采摘目标进行闭环定位,有效地解决了采摘机器人一次远景定位误差较大的问题。试验结果表明,排除温室复杂光照情况,机器人末端执行器定位精度达到2mm,满足采摘作业要求。     

考虑光照条件的蛋壳蛋心颜色识别方法

研究了机器视觉识别中光照条件对蛋壳蛋心颜色的影响;根据白壳与青壳蛋透光性的差异,对比分析了鸭蛋不同壳色在不同光照强度下的颜色信息,找出了白壳蛋与青壳蛋图像信息差异最大的参数以及对应的光照条件.同时,分析了与蛋黄颜色相关性较大的特征参数及光照条件.实验表明:光照条件的亮度等级为120时,鸭蛋图像的可分度最好,利用形心区域的尺分量来判断蛋壳的颜色,计算量减少,分类准确,易实现;白壳蛋在180亮度及青壳蛋在210亮度等级下,根据形心区域的H分量建立判别模型,预测精度高.     

基于模糊模式识别理论的富营养化评价模型及应用

依据水体富营养化及其程度属于模糊概念这一客观实际,基于模糊模式识别理论,以叶绿素a、总磷、总氮、透明度和高锰酸盐指数作为评价指标,尝试建立富营养化评价的模糊模式识别模型。并以南方某大型水库为例,按年、汛期和非汛期3个阶段开展了富营养化评价研究。评价结果表明2005-2008年年均富营养化特征值分别为2.829 0、2.717 7、2.844 9、2.535 4,汛期分别为2.913 8、2.731 4、2.971 7、2.444 7,非汛期分别为2.748 3、2.712 5、2.713 8、2.648 8。可见该水库正处于由中营养向富营养过渡阶段,实例同时也表明该方法用于水体富营养化评价是合理、可行的。