<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[nizaoshake的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/nizaoshake</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; nizaoshake]]></copyright><item><title><![CDATA[边缘模板匹配算法]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/156887479</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/156887479</guid><author>nizaoshake</author><pubDate>Tue, 13 Jan 2026 11:11:23 +0800</pubDate><description><![CDATA[本文介绍了工业视觉领域中三种常用的模板匹配算法：轮廓匹配、灰度匹配和边缘匹配。通过对比分析，指出边缘匹配结合了轮廓匹配的计算效率和灰度匹配的精度优势，同时具有更好的光照鲁棒性。重点阐述了基于金字塔采样的边缘匹配算法实现方案，包括模板训练阶段的多层级边缘信息提取与旋转预计算，以及匹配阶段从粗到精的多级搜索策略。详细说明了算法核心的数据结构设计、梯度计算方法和相似度度量公式，并提出了通过OpenMP并行计算优化性能的建议。该算法在保证匹配精度的同时有效降低了计算复杂度，适用于工业场景下的目标定位需求。]]></description><category></category></item><item><title><![CDATA[支持向量机SVM区分条码还是二维码]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/145590646</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/145590646</guid><author>nizaoshake</author><pubDate>Wed, 12 Feb 2025 13:50:18 +0800</pubDate><description><![CDATA[框架就用python的sklearn库，直接贴代码了。]]></description><category></category></item><item><title><![CDATA[电子放大倍率计算方法]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/144430250</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/144430250</guid><author>nizaoshake</author><pubDate>Thu, 12 Dec 2024 16:41:16 +0800</pubDate><description><![CDATA[显示器为1920*1080,21寸，宽477mm，2700像素的金手指区域所占显示尺寸为477*2700/1920≈670mm。比如照片上17.2mm的元器件所占像素为2700，照片解析度就是17.2/2700≈6.37μm/pix。简单点想就是，显示器的单显示点实际尺寸：477mm/1920=a;当然要算整体相对物方实物的放大倍率还要*本身镜头的放大倍率c。相机采集图像本身单像素代表尺寸：17.2mm/2700=b。电子放大倍率，即显示器的放大率，显示器对你图像的放大倍率。显示放大倍率：a/b。]]></description><category></category></item><item><title><![CDATA[OpenCV多通道图像合并为单通道]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/138621153</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/138621153</guid><author>nizaoshake</author><pubDate>Thu, 09 May 2024 16:56:04 +0800</pubDate><description><![CDATA[当然最后一个功能值的选择具体要看合并通道数量，以及输入图像通道的排列，我把值放在下面供参考。很简单，一句代码的事。]]></description><category></category></item><item><title><![CDATA[使用opencv仿射变换warpaffine旋转图像]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/137221957</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/137221957</guid><author>nizaoshake</author><pubDate>Mon, 01 Apr 2024 10:04:59 +0800</pubDate><description><![CDATA[使用opencv的仿射变换进行图像旋转并保留所有像素信息]]></description><category></category></item><item><title><![CDATA[opencv提取图像的环形区域]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/135608007</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/135608007</guid><author>nizaoshake</author><pubDate>Mon, 15 Jan 2024 18:19:55 +0800</pubDate><description><![CDATA[不废话了，直接上代码。]]></description><category></category></item><item><title><![CDATA[使用opencv在矩形区域内沿指定方向搜索直线边沿]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/135508194</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/135508194</guid><author>nizaoshake</author><pubDate>Wed, 10 Jan 2024 17:15:04 +0800</pubDate><description><![CDATA[一般来说，选择边沿点的方式有First（第一个有效点），Last（最后一个有效点），Best（最大梯度点）三种。4. 得到各个卡尺的边沿点后使用Ransac方法过滤异常点。1. 沿卡尺方向提取搜索区域。3. 提取图像滤波求梯度。3. 按照选择方式输出点。]]></description><category></category></item><item><title><![CDATA[使用opencv提取旋转矩形区域]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/135460609</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/135460609</guid><author>nizaoshake</author><pubDate>Mon, 08 Jan 2024 16:57:25 +0800</pubDate><description><![CDATA[opencv提取旋转矩形区域]]></description><category></category></item><item><title><![CDATA[计算向量的角度]]></title><link>https://blog.youkuaiyun.com/nizaoshake/article/details/134941392</link><guid>https://blog.youkuaiyun.com/nizaoshake/article/details/134941392</guid><author>nizaoshake</author><pubDate>Tue, 12 Dec 2023 09:54:34 +0800</pubDate><description><![CDATA[简化了向量围绕原点角度计算的函数]]></description><category></category></item></channel></rss>