3*3卷积核实例
2016-04-01 10:33
363 查看
(转)3x3 convolution kernels with online demo
Which kernel is used for averaging, applying blur or smooth effect, do sharpening or for the emboss effect?
Which kernels can be used to detect edges, calculate the gradient or the smoothed gradient?
Can I try somewhere the two dimensional convolution in an interactive application?
The frequently used 3x3 convolution kernels are listed below with some short description. At the end of this post there is an interactive demo, where you can try and play with different 3x3 kernels. For the mathematical background of the usage of these kernels, please read the previous post on two dimensional convolution.
This kernel is used for noise reduction and blurring the image. Must be normalized, otherwise the result may not fit the (0, 255) range.
This kernel is used to enhance the small differences and edges in the image.
These kernels are sensitive to the edges. Kernel E is for detecting in both directions, while EH and EV are sensitive for the horizontal and vertical edges respectively.
Kernels GH and GV are to calculate the magnitude of the horizontal and the vertical gradient.
Sobel operators are similar to the gradient kernels approximating the smoothed gradient of the image in horizontal and vertical directions. It can be seen from the decomposing, that this operator is a combination of a gradient detector and a smoothing kernel.
This kernel creates an embossing effect, can be rotated to modify the direction of this operator.
Demo APP见原文
You can use the currently filtered image as source by clicking the Use filtered image button. Some predefined kernels can be chosen, but values are directly selectable too. Setting Filter normalization divides the convolution result by the summary of the elements in the kernel. It is important to mention, that on the result image the absolute output values are shown.
3x3 convolution kernels with online demo
Which are the most used 3x3 convolution kernels/matrices?Which kernel is used for averaging, applying blur or smooth effect, do sharpening or for the emboss effect?
Which kernels can be used to detect edges, calculate the gradient or the smoothed gradient?
Can I try somewhere the two dimensional convolution in an interactive application?
The frequently used 3x3 convolution kernels are listed below with some short description. At the end of this post there is an interactive demo, where you can try and play with different 3x3 kernels. For the mathematical background of the usage of these kernels, please read the previous post on two dimensional convolution.
Average (blur, smooth) 3x3 convolution kernel
This kernel is used for noise reduction and blurring the image. Must be normalized, otherwise the result may not fit the (0, 255) range.
Sharpen 3x3 convolution kernel
This kernel is used to enhance the small differences and edges in the image.
Edge detection 3x3 convolution kernels
These kernels are sensitive to the edges. Kernel E is for detecting in both directions, while EH and EV are sensitive for the horizontal and vertical edges respectively.
Gradient detection 3x3 convolution kernels
Kernels GH and GV are to calculate the magnitude of the horizontal and the vertical gradient.
Sobel operator 3x3 convolution kernels
Sobel operators are similar to the gradient kernels approximating the smoothed gradient of the image in horizontal and vertical directions. It can be seen from the decomposing, that this operator is a combination of a gradient detector and a smoothing kernel.
Emboss 3x3 convolution kernel
This kernel creates an embossing effect, can be rotated to modify the direction of this operator.
Demo application
This program demonstrates using 3x3 convolution kernels on classic image processing source images. Please click on the label to load the application. It shall run in every modern browser, including IE9+.Demo APP见原文
You can use the currently filtered image as source by clicking the Use filtered image button. Some predefined kernels can be chosen, but values are directly selectable too. Setting Filter normalization divides the convolution result by the summary of the elements in the kernel. It is important to mention, that on the result image the absolute output values are shown.
相关文章推荐
- 数据结构 顺序串的各种基本运算
- 当提示找不到某个文件路径的时候
- 深搜+回溯 poj3009 Curling 2.0
- Unity中实现高效Signal-slot模式--高效消息分发
- ./ggsci: error while loading shared libraries: libnnz11.so: cannot open shared object file: No such
- 菜鸟的Android学习笔记(二)
- Flex 布局教程:语法篇
- matlab的二维卷积操作
- 如何分析matlab程序的主要效率问题
- 安装Java
- JDK的帮助文档
- 第一个deeplearning4jproject跑起
- (转)Eclipse 下找不到或无法加载主类的解决办法
- PyCharm使用技巧记录(一)如何查看变量
- 到底该如何入门Keras、Theano呢?(浅谈)
- Keras如何构造简单的CNN网络
- PyCharm如何设置显示行号?
- Python/Keras如何将给定的数据集打乱
- Python如何进行cross validation training
- Python常用函数