Description: 基于GMM的概率神经网络PNN具有良好的泛化能力，快速的学习能力，易于在线更新,并具有统计学的贝叶斯估计理论基础，已成为一种解决像说话人识别、文字识别、医疗图像识别、卫星云图识别等许多实际困难分类问题的很有效的工具。而且PNN不但具有GMM的大部分优点，还具有许多GMM没有的优点，如强鲁棒性，需要更少的训练语料，可以和其他网络其他理论无缝整合等。-GMM based probabilistic neural network PNN good generalization ability, the ability to learn fast, easy online updates, and with the Bayesian statistical theory based on estimates, and has become a solution as speaker recognition, text recognition, medical image recognition, satellite images and other real recognition when difficulties classification of very effective tool. But GMM PNN is not only the most advantages, but also has many advantages GMM not as strong robustness, require less training corpus, and other networks to other theories, such as seamless integration. Platform: |
Size: 7158 |
Author:姜正茂 |
Hits:
Description: 基于GMM的概率神经网络PNN具有良好的泛化能力，快速的学习能力，易于在线更新,并具有统计学的贝叶斯估计理论基础，已成为一种解决像说话人识别、文字识别、医疗图像识别、卫星云图识别等许多实际困难分类问题的很有效的工具。而且PNN不但具有GMM的大部分优点，还具有许多GMM没有的优点，如强鲁棒性，需要更少的训练语料，可以和其他网络其他理论无缝整合等。-GMM based probabilistic neural network PNN good generalization ability, the ability to learn fast, easy online updates, and with the Bayesian statistical theory based on estimates, and has become a solution as speaker recognition, text recognition, medical image recognition, satellite images and other real recognition when difficulties classification of very effective tool. But GMM PNN is not only the most advantages, but also has many advantages GMM not as strong robustness, require less training corpus, and other networks to other theories, such as seamless integration. Platform: |
Size: 7168 |
Author:姜正茂 |
Hits:
Description: 基于动态轮廓模型的虹膜定位，在数字图像中虹膜位置的有效定位是虹膜识别的关键问题，与常见的定位方法相比，文中的方法速度快、精度高，而且，对瞳孔初始的伪圆心要求不高，鲁棒性更强-Contour model based on dynamic positioning of the iris, in the digital image of the iris position Effective positioning is the key to iris recognition problem, with the common positioning methods, the text of the method is fast, high precision, but also on the initial pseudo-pupil center do not ask for much, more robust Platform: |
Size: 270336 |
Author:姜浩 |
Hits:
Description: 图像识别，文字识别，手写输入识别，还包括打印处理等与图像相关处理方法-Image recognition, text recognition, handwriting recognition, but also deal with, such as print and image-related treatment methods Platform: |
Size: 13591552 |
Author:汪国梁 |
Hits:
Description: 一个图像文字识别软件源码，可以识别给定图像上的文字并导出文档-An image source text recognition software to identify a given text and images on export documentation Platform: |
Size: 3349504 |
Author:fivemice |
Hits:
Description: 索书号文字图像分割、粘连字符切分、文字识别、彩色车牌分割-Call Number text image segmentation, adhesion character segmentation, character recognition, color plate segmentation Platform: |
Size: 5120 |
Author:李荣华 |
Hits:
Description: 形码生成与识别的研究及实现(要求用C#或VB.NET实现 )1.条形码生成：通过对条形码结构、编码规则的学习和对C#中图像的读取及图像的基本操作的掌握，学会用GDI生成Code39条形码，或者用特殊字体的方式生成条形码，最终将输入的文字信息转化成符合国际标准的条形码。
2、条形码识别：确定系统框架，用C#完成课题的详细设计和代码编写，根据条形码的码制规则，对以图片方式获得的条形码通过图像处理相关知识经滤波、边缘检测、区域分割等处理进行识别，将条形码信息解析成相应文字-Code generation and recognition of research and the realization (requires C# Or VB.NET to achieve) 1. Bar code generation: structure of bar codes, coding rules for learning and C# Read in the image and the image of the basic operation of the master , learn how to use GDI to generate Code39 barcodes or special way to generate bar code fonts, the text will eventually enter the information into the bar code in line with international standards. 2, bar code recognition: the system framework, the C# Completed the detailed design and coding, the code system in accordance with the rules of the bar code on the way to the bar code image through image processing-related knowledge by filtering, edge detection, region segmentation, such as dealing with identification, bar code information into the appropriate text analysis Platform: |
Size: 12288 |
Author:wangchuanmeng |
Hits:
Description: 1.条形码生成：通过对条形码结构、编码规则的学习和对C#中图像的读取及图像的基本操作的掌握，学会用GDI生成Code39条形码，或者用特殊字体的方式生成条形码，最终将输入的文字信息转化成符合国际标准的条形码。
2、条形码识别：确定系统框架，用C#完成课题的详细设计和代码编写，根据条形码的码制规则，对以图片方式获得的条形码通过图像处理相关知识经滤波、边缘检测、区域分割等处理进行识别，将条形码信息解析成相应文-1. Bar code generation: structure of bar codes, coding rules for learning and C# Read in the image and the image of the basic operation of the master, learn how to use GDI to generate Code39 barcodes or special way to generate bar code font, will eventually enter text information into the bar code in line with international standards. 2, bar code recognition: the system framework, the C# Completed the detailed design and coding, the code system in accordance with the rules of the bar code on the way to the bar code image through image processing-related knowledge by filtering, edge detection, region segmentation, such as dealing with identification, bar code information into the appropriate text analysis Platform: |
Size: 3055616 |
Author:wangchuanmeng |
Hits:
Description: 1.条形码生成：通过对条形码结构、编码规则的学习和对C#中图像的读取及图像的基本操作的掌握，学会用GDI生成Code39条形码，或者用特殊字体的方式生成条形码，最终将输入的文字信息转化成符合国际标准的条形码。
2、条形码识别：确定系统框架，用C#完成课题的详细设计和代码编写，根据条形码的码制规则，对以图片方式获得的条形码通过图像处理相关知识经滤波、边缘检测、区域分割等处理进行识别，将条形码信息解析成相应文-1. Bar code generation: structure of bar codes, coding rules for learning and C# Read in the image and the image of the basic operation of the master, learn how to use GDI to generate Code39 barcodes or special way to generate bar code font, will eventually enter text information into the bar code in line with international standards. 2, bar code recognition: the system framework, the C# Completed the detailed design and coding, the code system in accordance with the rules of the bar code on the way to the bar code image through image processing-related knowledge by filtering, edge detection, region segmentation, such as dealing with identification, bar code information into the appropriate text analysis Platform: |
Size: 11264 |
Author:wangchuanmeng |
Hits:
Description: GOCR is an OCR (Optical Character Recognition) program, developed under the GNU Public License. It converts scanned images of text back to text files. Joerg Schulenburg started the program, and now leads a team of developers.
GOCR can be used with different front-ends, which makes it very easy to port to different OSes and architectures. It can open many different image formats, and its quality have been improving in a daily basis. Platform: |
Size: 382976 |
Author:Yao |
Hits:
Description: 内含《Visual C++ MATLAB图像处理与识别实用案例精选》pdg格式（用超星阅读器软件打开）的教材内容和用matlab编写源码，以及一些数字、字母等待识别的图片。非扫描格式，文字可以进行复制、粘贴操作。
内含一下几种格式文件：*.jpg、*.m、*.cpp、*.h等-Includes " Visual C++ MATLAB image processing and recognition of practical cases selected" pdg format (with Superstar Reader software to open) the teaching materials and the preparation of source code with matlab, and some numbers, letters, pictures awaiting identification. Non-scan format, text can be copy, paste operations. Contains about several format:*. jpg,*. m,*. cpp,*. h, etc. Platform: |
Size: 14315520 |
Author:rengt |
Hits:
Description: ：对乐谱识别中的音符图像分割技术进行了研究，采用游程分析与二次投影技术对乐谱图像进行处理。
文中结合乐谱的语法特征对乐谱图像进行分析，先对图像中的音乐音符进行初步分类，对每一类的音符采用不
同的处理方法，从而解决了乐符分割的难题。 -: Notes on the music recognition image segmentation techniques were studied, using run-length analysis and re-projection image processing technology to music. The syntax of the text with music feature on the music image analysis, the first image in the initial classification of musical notes, for each class of notes with different approaches to solve the problem of the music division operator. Platform: |
Size: 702464 |
Author:aaww |
Hits:
Description: Description:
A Java/.NET GUI frontend for Tesseract OCR engine. Supports optical character recognition for Vietnamese and other languages supported by Tesseract.
VietOCR is released and distributed under the Apache License, v2.0.
Features:
Multi-platform (Java version only)
Windows
Solaris
Linux/Unix
Mac OS X
Others
PDF, TIFF, JPEG, GIF, PNG, BMP image formats
Multi-page TIFF images
Screenshots
Selection box
File drag-and-drop
Paste image from clipboard
Postprocessing for Vietnamese to boost accuracy rate
Vietnamese input methods
Localized user interface
Integrated scanning support (on Windows only)
Watch folder monitor for support of batch processing
Custom text replacement in postprocessing
Spellcheck with Hunspell
Support for downloading and installing language data packs and appropriate spell dictionaries
Bravenet Counter Stats-Description:
A Java/.NET GUI frontend for Tesseract OCR engine. Supports optical character recognition for Vietnamese and other languages supported by Tesseract.
VietOCR is released and distributed under the Apache License, v2.0.
Features:
Multi-platform (Java version only)
Windows
Solaris
Linux/Unix
Mac OS X
Others
PDF, TIFF, JPEG, GIF, PNG, BMP image formats
Multi-page TIFF images
Screenshots
Selection box
File drag-and-drop
Paste image from clipboard
Postprocessing for Vietnamese to boost accuracy rate
Vietnamese input methods
Localized user interface
Integrated scanning support (on Windows only)
Watch folder monitor for support of batch processing
Custom text replacement in postprocessing
Spellcheck with Hunspell
Support for downloading and installing language data packs and appropriate spell dictionaries
Bravenet Counter Stats Platform: |
Size: 7952384 |
Author:BoinK |
Hits:
Description: 文字识别是图像识别的一部分，具有广泛的价值，本程序设计一个手写的数字识别功能-Text recognition is part of image recognition, with a wide range of value, the program design features a handwritten numeral recognition Platform: |
Size: 311296 |
Author:孙渊 |
Hits:
Description: 本书从机器学习的角度介绍了基于视觉的运动分析领域的最新算法和系统。-Techniques of vision-based motion analysis aim to detect, track, identify, and generally understand the behavior of objects in image sequences. With the growth of video data in a wide range of applications from visual surveillance to human-machine interfaces, the ability to automatically analyze and understand object motions from video footage is of increasing importance. Among the latest developments in this field is the application of statistical machine learning algorithms for object tracking, activity modeling, and recognition.
Developed from expert contributions to the first and second International Workshop on Machine Learning for Vision-Based Motion Analysis, this important text/reference highlights the latest algorithms and systems for robust and effective vision-based motion understanding from a machine learning perspective. Highlighting the benefits of collaboration between the communities of object motion understanding and machine learning, the book discusses the most act Platform: |
Size: 10570752 |
Author:kankan |
Hits:
Description: A PhD thesis on Semi-supervised learning with Graphs by Xiaojin Zhu. Focuses on creating graphs, based on a mixture of labeled and unlabeled data (as per the semi-supervised learning paradigm) and using processes on these graphs to propagate in rigorous way the labels from the labeled data to the unlabeled.
Applications in various areas in the Machine learning, such as text processing, image processing, pattern recognition, and others. Platform: |
Size: 2542592 |
Author:stefan.petrov.beb |
Hits:
Description: 用directshow写视频获取图片抓拍和显示程序，有分辨率设置，文字信息加入视频和抓拍的图片，设置文字信息的颜色和大小,这里给出单行，自己可以改成多行或者鼠标定点加入文字信息,其中也用到GDI+对多种图片格式的显示和保存。因在VC2010编译出的，里面包含已编译好的安装程序，此程序抓取的图片也可作为字符文字识别的测试平台，方便测试自己写的文字识别函数代码-For image capture and display program with the directshow write video resolution settings, text messages, video and snapping pictures, set the text color and size of the information given here to a single line, they can be changed to a multi-line text message or mouse sentinel join , which is also used to display and save GDI+ variety of image formats. VC2010 compiler, which contains the installation program has been compiled, the picture crawl in this program as a character text recognition test platform, to facilitate testing their own code written character recognition function Platform: |
Size: 5832704 |
Author:曾祥园 |
Hits:
Description: 用VC_和MODI实现图像文字识别 -VC_ of image and text recognition with MODIVC_ of image and text recognition with MODI Platform: |
Size: 247808 |
Author:宋格格 |
Hits:
Description: 项目基于Tensorflow进行实现。
#### 文件说明：
---
* input_data.py: 图片的输入
* traffic_sign_cnn.py: 用cnn进行训练分类
* testDemo.py: 用于测试已经训练出来的模型，输入单个图片输出结果，并分类到文件夹
#### 数据集说明：
---
* 这里是列表文本使用的是比利时的交通标志数据集，可以网上自己找，里面有62个分类。
#### 网络说明：
---
* 这里是列表文本这里是列表文本CNN网络包含两个卷积层，两个全连接层。识别率大概在95% 左右，可以自己根据需要自己修改参数提高识别率
另外，训练开始前需要先在项目目录下新建文件夹./log/train/,用来保存模型参数,数据集的目录结构大概是./data/train/00001（标签）/图片(The project is based on Tensorflow.
#### File Description:
---
* input_data.py: input of the picture
* traffic_sign_cnn.py: Training classification with cnn
* testDemo.py: Used to test the trained model, enter the result of a single image output, and categorize it into a folder
#### Data Set Description:
---
* Here is a list of texts using the Belgian traffic sign data set, which can be found online, with 62 categories.
#### Network Description:
---
* Here is the list text Here is the list text The CNN network consists of two convolutional layers, two fully connected layers. The recognition rate is about 95%, you can modify the parameters by yourself to improve the recognition rate
In addition, before the start of training, you need to create a new folder in the project directory. / Log / train /, used to save the model parameters, the directory structure of the data set is probably ./data/train/00001 (tag) / picture) Platform: |
Size: 430080 |
Author:lionkiss |
Hits:
Search - image with text recognition