The use of color recognition to maintain a state matrix, after each step away from the state matrix and color information can get the information of current chessboard, and restore the state matrix. 使用颜色识别维护一个棋子信息的状态矩阵,由每一步走棋后的颜色信息和走棋前的状态矩阵可以得到当前的棋盘信息,从而更新状态矩阵。
Time-frequency Spectra Recognition Based on Sparse Non-negative Matrix Factorization and Support Vector Machine 基于稀疏性非负矩阵分解和支持向量机的时频图像识别
Palmprint Recognition Based on Non-Negative Matrix Factorization and General Discriminant Analysis 基于非负矩阵分解和广义判别分析的掌纹识别
Irregular-Closed Graphic Recognition Based on Contour Measurement Matrix 基于等高测度矩阵辨识不规则封闭图形
Independent component analysis for face recognition based on two dimension image matrix 基于二维图像矩阵的ICA人脸识别
Incremental learning of face recognition based on block non-negative matrix factorization 基于分块非负矩阵分解人脸识别增量学习
Facial Expression Recognition Research Based on Non-negative Matrix Factorization 基于非负矩阵分解的面部表情识别研究
Study and Implementation of Recognition Algorithm for Data Matrix Two-dimensional Barcode Image DataMatrix二维条码图像识别的算法研究与实现
Speech Recognition using Transfer Probability Matrix of Rise-Fall period 升降周期转移概率矩阵(TPMRF)语音识别方法的研究
The algorithm uses geometric information to merge the strokes into character blocks based on handwriting rules and applies dynamic programming to find the best segmentation path. The width-height ratio, size, inter-character distance, and recognition information are used to construct the cost matrix. 通过基于规则的笔划合并将单个笔划合并成字符块,在此基础上,采用动态规划算法,利用字符块的长宽比、大小、间距以及识别信息,寻找出最优的切分路径。
A speech recognition method based on transfer probability matrix of rise-fall period is described in this report. 通过对语音时间波形的分析和研究,我们选用语音波形的升降周期为特征,以其转移概率矩阵为模板,建立了新的识别算法。该方法数学公式简单,计算量小。
Distortion-invariant optical pattern recognition by correlative matrix feature criterion 用相关矩阵特征判据法实现三重不变光学图像识别
According to analysis and comparison to the fuzzy cluster theory and methods, we believe that fuzzy C-cluster is a good method to fault recognition. Using it, we got best-classified matrix and center matrix, which could construct for four kinds of normal fault models. 通过对模糊聚类理论方法的分析比较,引入模糊C聚类划分理论及方法对噪声信号的指标样本进行分类,得到最优分类矩阵和聚类中心,从而建立了故障的标准类型样本。
Recognition effect of Zernike matrix is much better than space geometry matrix is proved from theory and experiment. 并从理论和实验两个方面证实Zernike矩的识别效果优于空间几何矩。
Time-Space Data Fusion and Object Recognition Based on Matrix Analysis and Dempster-Shafer Evidence Theory 基于矩阵分析和D-S证据理论的时空数据融合及目标识别
This paper introduces the self-educated recognition matrix of using image recognition method in various geophysical data for the order of classifying on the research geologic objective and resolving the problem of drawing geologic map and so on, and it discusses the matrix's principle and algorithm. 介绍了利用图像识别方法对各种地球物理资料进行综合识别以达到对所研究地质目标进行统计分类及解决地质填图等目的的自学识别模块,就其方法原理和算法进行了详细的论述。
A Method of Pattern Recognition Based on Associative Matrix 一个基于相关存储的模式识别方法
Pattern recognition theory comes down to algebra, matrix theory, probability theory, graph theory, fuzzing mathematics and optimization theory. 模式识别是研究分类识别理论和方法的科学,在理论上涉及到代数学、矩阵论、概率论、图论、模糊数学、最优化理论等等众多学科的知识,其内涵可概括为信息处理、分析、决策。
Novel Method for Printed Character Recognition: ( 0,1,)-Matrix Based on GA 一种印刷体字符识别的新方法:基于遗传算法的(0,1,)-矩阵法
The Pattern Recognition Method for Correcting the Judgement Matrix into One with Complete Uniformity in AHP AHP中判断矩阵一致性修正的模式识别法
The experimental results indicate that high recognition rate can be obtained through the appropriate selection of the dimension of block matrix although there exists nonlinear relationship between recognition rate and dimension of block matrix. 实验结果表明,虽然识别率与分块维数之间存在非线性关系,但可以通过选择适当的分块维数来获得较高的识别率。
A new method used for automatic detection and classing of thermal relays was introduced. The simple and practical algorithm was obtained on the basis of fuzzy pattern recognition theory by establishing optimum opposite membership of degree matrix of standard special characteristic and optimum fuzzy pattern recognition matrix. 介绍一种检测热继电器的特性曲线和分档的新方法,其特点是基于模糊模式识别理论,通过建立最优标准相对隶属度矩阵和最优模糊模式识别矩阵,得到概念清晰、简洁实用的算法。
Recognition of Hepatic Carcinoma Sonograph Using Gradient Differential Matrix Method 梯度差分矩阵法在肝癌声象图识别中的应用
This paper verifies modal curvature difference method in theory at first, aiming at the feasibility of the damage recognition of structures. A matrix formula by using modal vector difference before and after damage is used to evaluate the damage factors of structures. 首先在理论上证明模态曲率差法,对结构损伤识别的可行性,提出了用损伤前后的模态矢量差,来计算结构损伤因子矩阵的计算公式。
The investigation of image recognition based on matrix computing. 第二部分基于矩阵计算的理论和方法,研究了图像识别问题。
In the feature extraction process, a new face recognition method based on CSVD and non negative matrix factorization ( NMF) is presented. 在人脸识别特征提取环节,提出了CSVD算法与非负矩阵因子(NMF)算法特征数据相融合的人脸识别算法。
This thesis researches on the image pre-processing techniques involved in automatic recognition of data matrix barcode captured by camera as an example. 本文以DataMatrix(DM)二维条形码为例,研究基于摄像头的二维条码图像自动识别所涉及到的图像预处理技术。
In this paper, we reviewed and summarized the research on earthquakes and explosions identification, studied the rapid identification criteria between earthquakes and explosions and multiple characteristics recognition by using pattern recognition methods and matrix decision-making algorithm ( MDA). 本文对地震和爆破的识别研究进行了回顾和总结,并研究了地震和爆破的快速识别判据以及用模式识别方法和决策方法进行多特征综合识别的方法,取得较好的识别结果。
The principal component analysis recognition algorithm with intraclass scatter matrix as the generating matrix is designed and implemented. 设计实现了以类间散步矩阵为产生矩阵的主成分分析字符识别算法。
Molecularly imprinted technology ( MIT) is an efficient method to introduce specific molecular recognition sites into a polymeric matrix. 分子烙印技术(MIT)是一种将特定的分子识别位点引入聚合物基质的有效方法。