tensorflow test

连接边缘算法提取的孤立点，实现图像边缘连接(link the point to get edges in the image )

本文提出了一种基于数字图像轮廓特征的机械零件自动有 效识别的系统。利用工业摄像镜头代替目视作为一个传感器，经过光电转换、图 像采样、图像处理、图像分析识别等一系列的操作，就可以达到机械零件识别的 目的。(This paper presents a digital image based on the outline of the characteristics of mechanical parts automatically effective identification system. Utilization of industrial camera lens as a substitute for visual sensors, through photoelectric conversion, image sampling, image processing, image analysis to identify a series of operations to achieve the purpose of identification of mechanical parts.)

高阶频谱分析工具箱，老外写的代码，可以用来做实验(High-order spectral analysis toolbox, a foreigner writing code that can be used to experiment)

模糊集图像增强：设计隶属度函数将图像从空间域变换到模糊集域；设计模糊增强算子，在模糊集域对图像进行处理；根据隶属度函数重新将图像从模糊集域变换到空间域。(Fuzzy image enhancement: Designing the membership function of the image from the spatial domain to the fuzzy set domain design fuzzy enhancement operator in fuzzy set domain for image processing membership function according to re-image from the fuzzy set domain to the spatial domain .)

高斯混合模型的最大期望迭代求解算法,可用于图像区域灰度分布估计(Expectation maximazation(EM) for Gaussian mixture model(GMM) )

说明：  八值图像连通区域标记，为每个连通区域分配一个唯一的标号，处理后的图像按照从左到右，从上到下的顺序获得连续的标号(Binary image connected region eight marks, for each connected region a unique label distribution deal with in accordance with the image from left to right, from top to bottom order of access to continuous labeling)

该实例实现了 截图并识别截图中的文字功能 该项目需要在安装 Microsoft Office Document Imaging 下载地址 如下: http://www.xdowns.com/soft/4/25/2009/Soft_50716.html 

混合高斯模型提取视频帧前景背景图像，经试验验证准确率比较高(Mixed gaussian model to extract video frame prospects background image)

用小波包提取故障特征向量并归一化，很实用，论文里的部分程序(Extracted using wavelet packet fault eigenvector and normalized, it is useful, some of the procedures in the thesis)

给定任意一个子波，通过克基霍夫变换返回最小相位子波。(Kolmogoroff spectral factorization. Given an arbitrary wavelet this function retrieves the minimum phase wavelet using Kolmogoroff factorization)