Pd Pattern Recognition Based on Linear Discriminant Analysis in GIS

According to the character of PD in GIS, the authors design four kinds of GIS defection models. The GIS gray intensity images are constructed based on mass specimens gathered by the ultra - high frequency and high speeds systems, Aiming at the PD characteristics and its defections, A PCA-FDA method is put forward based on PD images. The principal component analysis is employed to condense the dimension of PD images, then the optimal sets of statistically uncorrelated discriminant vectors are extracted, and the minimum distance classifier is constructed as classifier. The identified results show that this method can effectively elevated the discrimination of the four kinds of defects in GIS PD.     

Developing a Recognition System of Gun-code

Intelligent recognition system of gun code is designed , which achieves online recognition of gun code. We applied video camera to get image, and some special technique in this deformed character recognition. As well, we applied dynamic template matching to get a better result.     

Detecting Technology of Cabel Surface Defect Based on DSP and Image Recognition

The damage of bridge cable directly hazards the normal use of bridge and the safety of people. The paper discussed the signal acquisition and detection system which takes embedded DSP processor as core technology, combined with CCD, video decode chip, SDRAM, FLASH ROM, and established the fast measuring algorithm of the surface corrosion and damage, which includes deteminating the processing region, improving medium filter, sobel edge detection and dynamic threshold segment, close operation of morphology, computing defect area and saving. The experimental resuits show that this method implemented the real-time, non-contact and intelligence detecting of cable surface defect and gives the specifications of the system.     

一种快速判别梨果梗的方法

果梗完好与否是梨品质检测的指标之一,因而对果梗情况进行准确判别具有重要意义。为此,通过计算机视觉系统摄取梨的图像,利用图像处理技术提出了一种能快速判别梨果梗有无的算法。该算法的识别正确率达90%,识别速度大约在20～30ms,实现了对梨果梗进行高速检测的目标。     

采棉机器人的研究现状及关键技术

对国内外机械化采棉技术的研究现状进行了综述,构建出了基于计算机视觉的智能型采棉机器人原型系统,分析了采棉机器人的应用价值.同时,介绍了采棉机器人的作业环境和工作对象的特殊性,总结了采棉机器人所涉及的关键技术,包括采棉机器人本体的优化设计、棉花的自动识别和分类以及路径规划和运动控制技术等,最后指出了目前应用采棉机器人存在的主要问题并提出相应对策.     

属性识别模型在水利水电规划方案综合优选中的应用

针对目前水利水电规划方案综合优选方法中存在的问题，提出了基于变异系数权重的水利水电规划方案综合优选属性识别模型。该模型采用反映评价指标特征值之间差异性的参数——变异系数作为评价指标的权重，减小了权重确定的主观性，且模型求解方法简单易学。实例分析表明，这一模型和方法是合理、可行的。     

增压器转速适应性测量方法

提出一种以采集卡为测量平台的增压器转速测量方法。系统以软件方式实现了随转速变化进行自适应调整的数字滤波，给出一种基于转速信号极值点的周期识别方法。分析和实验结果表明，该方法能够在增压器转速信号受外界干扰而偏离的情况下，在较宽的范围内实时测量增压器转速。     

基于双池化与多尺度核特征加权CNN的典型牧草识别

针对自然背景下牧草难识别的问题,提出一种基于双池化与多尺度核特征加权的卷积神经网络牧草识别方法。双池化特征加权结构通过将卷积层输出的特征图分别进行最大值池化和均值池化得到两组特征图,引入特征重标定策略,依照各通道特征图对当前任务的重要程度进行加权,以增强有用特征、抑制无用特征;多尺度核特征加权结构通过在卷积层中同时使用3×3和5×5两种卷积核,并将网络的前几层特征复用后进行加权,以提高重要特征的利用率。对10类牧草图像进行识别实验,结果表明,该方法识别率为94. 1%,比VGG-13网络提高了5. 7个百分点,双池化与多尺度特征加权有效提高了牧草识别精度。