搜索资源列表
classify
- SVM的数据分类预测——意大利葡萄酒种类识别-SVM prediction data classification- Italian wines Recognition
HMM-C-based
- HMM基于C语言的模型程序源代码，语音识别，数字信号处理-HMM model based on the C language source code, voice recognition, digital signal processing
SpeakerDemoCode
- speech recognition source code
ECR_R
- 利用小波变换提取脑电中的R波~识别率可达百分之九十五以上~并附带脑电数据！-EEG using wavelet transform to extract the R wave ~ recognition rate of more than 95 percent- and comes with EEG data!
sgu262
- sgu262：Symbol Recognition(状压DP) 题目大意： 给出k个n∗ m的01矩阵Si，求出一个1尽量少的n∗ m的01矩阵P，满足k个矩阵与该矩阵的交互不相同，也就是说通过该矩阵能表示出给出的k个矩阵。-262. Symbol Recognition time limit per test: 0.25 sec. memory limit per test: 65536 KB input: standard output:
HanLP-1.2.7
- HanLP是一个致力于向生产环境普及NLP技术的开源Java工具包，支持中文分词（N-最短路分词、CRF分词、索引分词、用户自定义词典、词性标注），命名实体识别（中国人名、音译人名、日本人名、地名、实体机构名识别），关键词提取，自动摘要，短语提取，拼音转换，简繁转换，文本推荐，依存句法分析（MaxEnt依存句法分析、神经网络依存句法分析）。-HanLP is a dedicated to popularize NLP technology to production environment of
CCA
- 新型可信性聚类算法CCA，聚类性能比FCM和K-Means要好很多-A data clustering method partitions unlabeled data sets into clusters and labels them for various goals such as computer vision and pattern recognition.
narf_feature_extraction
- pcl point 选择点云数据中的关键点，对点云识别具有重要应用，值得学习- pcl point the point cloud data in the key points of the point cloud recognition has important applications, it is worth learning
dtw.m
- 语音识别的计算function文件。 可以计算出2个语音之间的差值并且无视语速的问题。 值越小说明相似度越高。 -Voice recognition function calculation file. You can calculate the difference between the two voice and speech rate of ignoring the problem. The smaller the value the higher the degree of simil
banhun
- 粒子图像分割及匹配均为自行编制的子例程，BP神经网络用于函数拟合与模式识别，采用偏最小二乘法。- Particle image segmentation and matching subroutines themselves are prepared, BP neural network function fitting and pattern recognition, Partial least squares method.
faofang
- 是信号处理的基础，LCMV优化设计阵列处理信号，通过反复训练模板能有较高的识别率。- Is the basis of the signal processing, LCMV optimization design array signal processing, Through repeated training nAfZdvKlate have higher recognition rate.
faohie
- 是一种双隐层反向传播神经网络，正确率可以达到98％，人脸识别中的光照处理方法。- Is a two hidden layer back propagation neural network, Accuracy can reach 98 , Face Recognition light treatment method.
fieyan_v42
- 完整的基于HMM的语音识别系统，包括随机梯度算法，相对梯度算法，滤波求和方式实现宽带波束形成。- Complete HMM-based speech recognition system, Including stochastic gradient algorithm, the relative gradient algorithm, Filtering summation way broadband beamforming.
goufui
- 具有丰富的参数选项,Gabor小波变换与PCA的人脸识别代码，通过反复训练模板能有较高的识别率。- It has a wealth of parameter options, Gabor wavelet transform and PCA face recognition code, Through repeated training vPAIMMQlate have higher recognition rate.
hiufang_v36
- 实现了对10个数字音的识别，利用自然梯度算法，ofdm系统仿真 含16qam调制 fft 加窗 加cp等模块。- To achieve the recognition of 10 digital sound, Use of natural gradient algorithm, ofdm system simulation including 16qam modulation fft windowing modules plus cp.
junnao
- 基于matlab GUI界面设计,实现了对10个数字音的识别，基于混沌的模拟退火算法。- Based on matlab GUI interface design, To achieve the recognition of 10 digital sound, Chaos-based simulated annealing algorithm.
kuinui_v10
- 通过反复训练模板能有较高的识别率，包括脚本文件和函数文件形式，实现用SDRAM运行nios，同时用SRAM保存摄像头数据。- Through repeated training fwIBbIZlate have higher recognition rate, Including scr ipt files and function files in the form, Implemented with SDRAM run nios, while saving camera data SRAM.
laonao_v42
- BP神经网络用于函数拟合与模式识别，Matlab实现界面友好，最大信噪比的独立分量分析算法。- BP neural network function fitting and pattern recognition, Matlab to achieve user-friendly, SNR largest independent component analysis algorithm.
km4
- 一种面向工艺规划的零件分类和识别方法：采用零件工艺相似系数对零件实现分类。类似K-means的聚类算法。-Part classification and recognition method for process planning: the classification of parts by the process similarity coefficient. A clustering algorithm similar to K-means.
muisan_v29
- BP神经网络用于函数拟合与模式识别，对HARQ系统的吞吐量分析，代码里有很完整的注释和解释。- BP neural network function fitting and pattern recognition, HARQ throughput analysis of the system, Code, there are very complete notes and explanation.