统计模式识别 meaning in English
statistic pattern recognition,
statistical pattern recognition
Examples
- The main contents of the thesis are shown as follows : presenting fundamental theories of statistic pattern recognition , discussing rgb ( red , green , blue ) color space , ohta color space , hsi ( hue , saturation , intensity ) color space and its converted color space , materials consistency in gray scale and the application in removing foreign bodies in tobacco flows , hence presenting recognition pattern based on " unit recognition " , designing sample machine for this purpose , which consists of material - providing system , optic system , image - grabbing system , real - time intelligent image - processing system and systems of automatically rejecting foreign bodies and self - diagnosis , analyzing and optimizing hard wares , offering concrete designs such as optic system and air - ejector driver circuit , presenting and realizing physical ram
本文的主要内容有：统计模式识别基础理论及它们在烟草异物识别中的应用；讨论了rgb （ red 、 green 、 blue ）基础颜色空间、 ohta颜色空间、 hsi （ hue色调， saturation饱和度， intensity亮度）颜色空间及其变换空间、物料图像纹理、灰度均匀性等在烟草异物识别中的应用，并在此基础上提出了"基于判别单元颜色统计特性"的烟草在线异物识别模型，设计并研制了烟草在线异物实时识别与自动剔除系统原理样机，它由供料系统、光学系统、图像数据采集系统、实时智能图像处理系统、异物自动剔除系统以及自诊断系统等组成。 - This recognition method integrates traits of nearest neighbor method and k - neighbor method . its basic principle is to improve recognition reliability with tow additive thresholds . in addition , ternary threshold neighbor recognition method ’ s fast arithmetic based on c - average clustering and huffman tree is put forward
2 .本文在经典统计模式识别方法的基础上,结合了最近邻识别法和k近邻识别法的特点,给出一种三阈值近邻识别方法,其基本原则是在寻找未知样本的近邻时,通过附加阈值的限制,进一步提高识别的可靠性。 - Currently , image recognition has five primary methods : classical statistic model method , auto method based on information , auto method based on model , amalgamation of several sensors and manual neural network . how to improve the classical method and how to assemble different recognition method are still problems to be improved . based on the current research state , the classical statistic model recognition method and correlative knowledge are researched firstly
基于当前的研究现状,本文以可见光和红外仪器同步拍摄的不同环境下的运动车辆目标序列为对象,在深入研究了经典的统计模式识别方法及相关理论的基础上,对车辆目标识别系统中的特征提取和识别方法等方面做了较深入的研究,并对神经网络在车型识别系统中的应用作了试探性研究。