ransetideshuliangtongji

给予分水岭的图像分割，包括源代码、测试图、实验报告(Given watershed image segmentation, including source code, test plans, test reports)

利用回转窑生产氧化铝的工艺中，提出了一种新的基于SVR（支持向量回归）的烧成带温度软测量方法。即首先对从回转窑采集到的烧结图像进行预处理，消除噪声的干扰；然后利用大津方法分割出物料区和火焰区，并提取这些区域的图像特征；其次利用基于SVR方法建立的烧成带温度模型计算温度值；参数计算.(The use of alumina rotary kiln production process, a new-based SVR (support vector regression) of the burning zone temperature soft measurement method. That is first collected from the rotary kiln sintering image pre-processing to eliminate noise interference Otsu method and then use the materials division of the district and flame zones, and extraction of image features in these regions SVR-based method followed by the establishment of the burning zone temperature model temperature parameters.)

％通过耦合神经网络结合中值滤波实现图像的降噪源代码( Through the combination of coupled neural network filtering image noise source code)

说明：  利用MATLAB对图像进行文件夹中图片进行批量增广(MATLAB is used to enlarge the images in the folder)

改进的 SUSAN 算法，用于自适应阈值的边缘检测(improved algorithm for the adaptive thresholding edge detection)

利用图像处理的方法实现图像二值化，实现了图像的调幅加网，加网角度是18.4度。这就方便了图像的调幅加网输出。(The use of image processing method of image binarization realize, realize the image of AM screening, screening angle is 18.4 degrees. This facilitates the image output of AM screening.)

基于SUFT的特征提取与匹配，内涵详细的文档说明，很好的实现了特征提取与匹配，该方法较基本的SIFT有很大的提高(Extraction and matching of features based on SUFT, detailed content of the documentation, very good realization of the feature extraction and matching, SIFT the method is basic is greatly improved)

说明：  图像特征提取的Contourlet变换工具包，可以设置分解层次，例如，三层分解提取17维特征，用于纹理图像或者SAR图像分割应用中(Image feature extraction toolbox of Contourlet transform, you can set the level of decomposition, for example, three level with 17-dimensional, for the segmentation on texture image or SAR Image)

卷积神经网络深度学习，可以识别多种图像，并自动解出所需要的结果(Deep learning of convolutional neural networks)

混合高斯模型使用K（基本为3到5个） 个高斯模型来表征图像中各个像素点的特征,在新一帧图像获得后更新混合高斯模型,用当前图像中的每个像素点与混合高斯模型匹配,如果成功则判定该点为背景点, 否则为前景点。通观整个高斯模型，他主要是有方差和均值两个参数决定，,对均值和方差的学习,采取不同的学习机制,将直接影响到模型的稳定性、精确性和收敛性。由于我们是对运动目标的背景提取建模，因此需要对高斯模型中方差和均值两个参数实时更新。为提高模型的学习能力,改进方法对均值和方差的更新采用不同的学习率 为提高在繁忙的场景下,大而慢的运动目标的检测效果,引入权值均值的概念,建立背景图像并实时更新,然后结合权值、权值均值和背景图像对像素点进行前景和背景的分类。(Gaussian mixture model using K (essentially 3-5) Gaussian model to characterize the features of each pixel in the image, in the image of the new frame for updated Gaussian mixture model, with each pixel in the image with a Gaussian mixture current model matching, if successful, determined that the point of the background points, otherwise the former attraction. Throughout the entire Gaussian model, he mainly has two parameters determine the variance and the mean, the mean and variance of the study, to take a different learning mechanism, will directly affect the stability, accuracy and convergence model. Since we are moving object extraction of the background modeling, so the need for the Gaussian model variance and mean two parameters real-time updates. In order to improve the learning ability of the model, an improved method for updating the mean and variance of different learning rates to improve in the busy scene, large and slow moving object detection results, the introduction of)