当前位置:网站首页>【图像分割】基于方向谷形检测实现静脉纹路分割附MATLAB代码
【图像分割】基于方向谷形检测实现静脉纹路分割附MATLAB代码
2022-07-28 21:24:00 【matlab_dingdang】
1 内容介绍
一种基于方向谷形检测的指静脉图像分割方法及其系统,终端,包括以下步骤:步骤1,接收图像采集模块采集的指静脉图像;步骤2,对截取的指静脉图像进行脊波变换作图像增强处理;步骤3,针对增强处理后的静脉图像提取谷形区域;步骤4,将得到的静脉图像分为三个区域:背景区,模糊区,前景区,并对该三个区域进行三次的阈值分割处理;步骤5,将得到的分割图像进行边缘检测,输出处理后的分割图像.本发明的系统及方法根据静脉图像中的蕴涵的一些内在信息如形状,阶跃性质,方向等特点而提出的一种算法,能够准确的提取图像中感兴趣的静脉特征,减少提取错误特征.
2 仿真代码
function shuchutuxiang=caijian(shurutuxiang,caijianshu)%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%输入参数shurutuxiang和caijianshu表示依输入二值图像和需要裁减的分支上的点数%输入参数表示shuchutuxiang裁减后的图像%思想是每次迭代除去端点直道有交叉点是停止。[m,n]=size(shurutuxiang);shurutuxiang=shurutuxiang;hh=zeros(m+2,n+2);canzhao=zeros(m+2,n+2);canzhao(2:m+1,2:n+1)=shurutuxiang(:,:);hh(2:m+1,2:n+1)=shurutuxiang(:,:);for l=1:caijianshufor i=2:m+1for j=2:n+1if hh(i,j)~=0&sum(sum(hh(i-1:i+1,j-1:j+1)))==2&sum(sum(canzhao(i-1:i+1,j-1:j+1)))<4shurutuxiang(i-1,j-1)=0;endendendhh(2:m+1,2:n+1)=shurutuxiang(:,:);endshuchutuxiang=shurutuxiang;
3 运行结果


4 参考文献
[1]龙安川, 唐忞旻, 曾尧,等. 基于方向谷形检测的指静脉图像分割方法及其系统,终端:, CN108010035A[P]. 2018.
[2]张锁平. 基于数学形态学图像分割的波向检测[J]. 海洋技术, 2007, 26(4):6.
博主简介:擅长智能优化算法、神经网络预测、信号处理、元胞自动机、图像处理、路径规划、无人机等多种领域的Matlab仿真,相关matlab代码问题可私信交流。
部分理论引用网络文献,若有侵权联系博主删除。
边栏推荐
- Gd32f303 firmware library development (10) -- dual ADC polling mode scanning multiple channels
- Record a question about the order of trigonometric function exchange integrals
- npm run dev,运行项目后自动打开浏览器
- Wheel 6: qserialport serial port data transceiver
- Seagate released a new risc-v architecture processor: the performance of mechanical hard disk soared 3 times
- CGLIb 创建代理
- Submission records of frontiers Publishing House (with status changes)
- Draem+sspcab [anomaly detection: block]
- In 2020, the top ten domestic IC design enterprises will be exposed! These five industrial challenges still need to be overcome!
- OSV-q The size of tensor a (3) must match the size of tensor b (320) at non-singleton dimension 3
猜你喜欢

es个人整理的相关面试题

console.log()控制台显示...解决办法

无代码开发平台管理后台入门教程

Target detection notes - overview and common data sets

How to delete and remove the first row of elements in PHP two-dimensional array

Yolov5 improvement 7: loss function improvement
![Fastflow [abnormal detection: normalizing flow]](/img/5e/984e5bd34c493039e3c9909fc4df05.png)
Fastflow [abnormal detection: normalizing flow]

Servlet的使用手把手教学(一)

Anomaly detection summary: intensity_ based/Normalizing Flow

《Shortening passengers’ travel time A dynamic metro train scheduling approach using deep reinforcem》
随机推荐
Xshell7, xftp7 personal free version official download, no need to crack, no activation, download and use
Yolov5 improvement 4: add ECA channel attention mechanism
一种分布式深度学习编程新范式:Global Tensor
Padim [anomaly detection: embedded based]
Console.log() console display... Solution
美国FCC提供16亿美元资助本国运营商移除华为和中兴设备
Labelme labels circular objects [tips]
一份来自奎哥的全新MPLS笔记,考IE必看 ----尚文网络奎哥
Yolov5 improvement 12: replace backbone network C3 with lightweight network shufflenetv2
How to delete and remove the first row of elements in PHP two-dimensional array
OSV_ Q write divergence operator div and Laplace stepped on the pit
The simple neural network model based on full connection layer MLP is changed to the model based on CNN convolutional neural network
OSV_ q Expected all tensors to be on the same device, but found at least two devices, cuda:0
Improvement 18 of yolov5: the loss function is improved to alpha IOU loss function
WebApplicationType#deduceFromClasspath
定了!哪吒S全系产品将于7月31日上市发售
Cglib create proxy
C语言学习内容总结
OSV_ q AttributeError: ‘numpy. ndarray‘ object has no attribute ‘clone‘
Evaluation index of anomaly detection: rocauc et al. [tips]