搜索资源列表
Digital-identification
- 进行数字识别,识别0~9的数字,识别率达到90 以上-Digital identification, identification of the number of 0~9, the recognition rate is above 90
project_code
- 基于神经系统的数字是识别,识别率一般。比较实用-Digital nervous system is based on recognition, the recognition rate in general. More practical
HandWriter
- 汉字手写识别apk, 一种手写软件算法库和测试apk,超高的识别率,可用于android平台的手写识别,可自动识别出简体中文和汉字等-HandWriter apk
handwriteREC
- 自己编写的一个matlab实现手写数字的识别,识别率高达90 以上。-I have written a matlab achieve recognition of handwritten digits recognition rate of 90 .
platerec
- 自己使用matlab编写的一个车牌识别程序,识别率百分之90以上。-Own use matlab license plate recognition program written in a recognition rate above 90 .
HLBP
- 基于Haar特征的LBP算法,在人脸识别方面会更加迅速,而且识别率很高-LBP algorithm based on Haar features in face recognition will be more rapid, and high recognition rate
cpdr-system
- 中文印刷体文档识别系统,基于VC++系统实现,可以准确识别中文印刷体,识别率高。-Chinese printed document recognition system, based on VC++ system implementation, can accurately identify the Chinese printing, high recognition rate.
Glpp-system
- 灰度车牌定位系统,基于VC++,可以识别灰度车牌,识别率不错。-Gray license plate positioning system, based on VC++, can identify the gray, the recognition rate.
PCA.m
- pca程序,经过测试,识别率在10 作用,效果良好。-pca program, tested, 10 percent recognition rate in effect, to good effect.
Grading-test
- 为实现合格和缺陷板栗的分级, 研究了 1 种基于 BP 神经网络与板栗图像特征的板栗分级方法。 试验以罗田板 栗为研究对象, 提取的颜色及纹理等 8 个特征值, 通过主成分分析提取相应的主成分得分向量构成模式识别的输入。 利 用 BP 神经网络方法建立了板栗分级模型。 试验结果表明, 在图像信息主成分因子数为 3, 中间层节点数为 12 时, 建立 的模型最佳, 模型训练时的回判率为 100 , 预测时识别率达到了 91 .
chepaishibie
- 实现蓝色车牌照的识别,车不能为蓝色,车牌倾斜方式有局限性只能为左右倾斜,识别率不是太高-Recognition achieve blue car license, vehicle can not be blue, there are limitations only way inclined plate is inclined about the recognition rate is not too high
face
- 人脸识别例程,识别率达到95 左右,有人脸库,下载可直接运行-Recognition routines, the recognition rate of around 95 percent, it was the face , download can be run directly
shuzishibie
- 数字识别系统源代码,完整的VC++程序,识别率在80 。-Digital recognition system source code, vc++ program, recognition rate at 80 .
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!
KL_SVD_face_recognition
- PCA主成分分析,采用KL投影和SVD分解提取人脸特征向量,最后采用最近邻判别法计算识别率。-Face recognition based on PCA. KL projection and SVD are used to extract face eigenvectors. Recognition rate is calculated by k nearest neighbors(KNN) method.
Face-recognition
- 人脸识别 识别率达90 opencv C语言-Face recognition rate of 90 opencv c
Face-recognition
- 基于BP神经网络的人脸识别MATLAB代码,构造、训练BP神经网络并计算其识别率。-Face recognition based on BP neural network MATLAB code, structure, training BP neural network and computing its recognition rate.
fruitvegtablerecognition_knn
- 基于KNN算法对多种水果和蔬菜图像进行分类识别,识别率较高。-Classifying a variety of fruits and vegetables images by using KNN algorithm.
fruitvegtablerecognition_svm
- 基于K-means算法和SVM算法对不同种类的水果蔬菜图像进行分类识别,算法识别率较高。- Classifying a variety of fruits and vegetables images by using K-means algorithm and SVM.
jianji
- 剪辑近邻法当不同类别的样本在分布上有交迭部分的,分类的错误率主要来自处于交迭 区的样本。当我们得到一个作为识别用的参考样本集时,由于不同类别交迭区域 中不同类别的样本彼此穿插,导致用近邻法分类出错。因此如果能将不同类别交 界处的样本以适当方式筛选,可以实现既减少样本数又提高正确识别率的双重目 的。为此可以利用现有样本集对其自身进行剪辑。 -Clips neighbor method when samples of d