OpenCv3.0+SVM的使用心得（一）

1.open cv3.0版本中没有了CvSVM类的定义，而是将其写入到一个document中，但是使用也并不复杂，代码示例如下：

<code class="hljs lasso has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;">ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::SVM</span><span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::Params</span> <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">params</span>;
<span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">params</span><span class="hljs-built_in" style="color: rgb(102, 0, 102); box-sizing: border-box;">.</span>svmType <span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">=</span> ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::SVM</span><span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::C_SVC</span>;
<span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">params</span><span class="hljs-built_in" style="color: rgb(102, 0, 102); box-sizing: border-box;">.</span>kernelType <span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">=</span> ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::SVM</span><span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::POLY</span>;
<span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">params</span><span class="hljs-built_in" style="color: rgb(102, 0, 102); box-sizing: border-box;">.</span>gamma <span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">=</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>;
Ptr<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;"><</span>ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::SVM</span><span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">></span> svm <span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">=</span> ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::SVM</span><span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::create</span>(<span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">params</span>);
Mat trainData; <span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">// 每行为一个样本</span>
Mat labels;    
svm<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">-></span>train( trainData , ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::ROW_SAMPLE</span> , labels );
<span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">// ...</span>
Mat query; <span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">// 输入, 1个通道</span>
Mat res;   <span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">// 输出</span>
svm<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">-></span>predict(query, res);</code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li><li style="box-sizing: border-box; padding: 0px 5px;">4</li><li style="box-sizing: border-box; padding: 0px 5px;">5</li><li style="box-sizing: border-box; padding: 0px 5px;">6</li><li style="box-sizing: border-box; padding: 0px 5px;">7</li><li style="box-sizing: border-box; padding: 0px 5px;">8</li><li style="box-sizing: border-box; padding: 0px 5px;">9</li><li style="box-sizing: border-box; padding: 0px 5px;">10</li><li style="box-sizing: border-box; padding: 0px 5px;">11</li><li style="box-sizing: border-box; padding: 0px 5px;">12</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li><li style="box-sizing: border-box; padding: 0px 5px;">4</li><li style="box-sizing: border-box; padding: 0px 5px;">5</li><li style="box-sizing: border-box; padding: 0px 5px;">6</li><li style="box-sizing: border-box; padding: 0px 5px;">7</li><li style="box-sizing: border-box; padding: 0px 5px;">8</li><li style="box-sizing: border-box; padding: 0px 5px;">9</li><li style="box-sizing: border-box; padding: 0px 5px;">10</li><li style="box-sizing: border-box; padding: 0px 5px;">11</li><li style="box-sizing: border-box; padding: 0px 5px;">12</li></ul>

另外，svm的定义和训练可以写到一起：

<code class="hljs lasso has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;">Ptr<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;"><</span>SVM<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">></span> svm<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">=</span>StatModel<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::train</span><span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;"><</span>SVM<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">></span>(trainData,ml<span class="hljs-tag" style="color: rgb(0, 102, 102); box-sizing: border-box;">::ROW_SAMPLE</span>,labels,<span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">params</span>)</code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul>

2.使用statModel::train(针对所有的机器学习 )的时候，要注意里面的参数（训练样本、训练标签）格式为InputArray，这个格式可以从Mat格式自动生成，但是不能从CvMat格式生成。

3.因此，必要的时候要进行两者的转换。当CvMat->Mat时：

<code class="hljs fix has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;"><span class="hljs-attribute" style="box-sizing: border-box;">Mat Matfeatures</span>=<span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">cv::cvarrToMat(CvMatFeatures)</span></code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul>

PS: 从IplImage格式转成Mat格式时也是调用此函数，从Mat转到IplImage格式时：

<code class="hljs fix has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;"><span class="hljs-attribute" style="box-sizing: border-box;">Mat mat</span>=<span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">&IplImage(img);//其中img是IplImage格式数据</span></code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul>

不知道这个方法对从Mat到CvMat的转换是否同样适用（估计是可以的）

4.由于3.0不存在为机器学习定义的类，因此不能用调用函数的方法对分类器进行操作，而代之使用:

<code class="hljs lasso has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;">svm<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">-></span>save(<span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">"...."</span>);<span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">//文件形式为xml，可以保存在txt或者xml文件中</span>
svm<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">-></span>getVarCount();<span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">//获取SVM支持向量的维数</span>
svm<span class="hljs-subst" style="color: rgb(0, 0, 0); box-sizing: border-box;">-></span>getSupportVectors;<span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;">//返回Mat类型，获取SVM的支持向量</span></code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li></ul>

5.可以对保存在xml或txt文件中的分类器进行加载。

<code class="hljs ruby has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;"><span class="hljs-constant" style="box-sizing: border-box;">Ptr</span><<span class="hljs-constant" style="box-sizing: border-box;">SVM</span>> svm=<span class="hljs-symbol" style="color: rgb(0, 102, 102); box-sizing: border-box;">statModel:</span><span class="hljs-symbol" style="color: rgb(0, 102, 102); box-sizing: border-box;">:load<SVM></span>(<span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">"...."</span>);</code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul>

6.在编写程序过程中，使用hog.compute()计算会报内存错误，折磨了一整天，最后将“项目属性”-“链接器”-“输入”中的”附加依赖项”中的ts300.lib和world300.lib删除掉了，只保留ts300d.lib和world300d.lib。具体原因，不详。

7.遇到”error LNK2019: unresolved external symbol __imp___CrtDbgReportW referenced in function “public: double & __thiscall std::vector…..”这种错误时，需要将” C/C++”->”代码生成”的运行库由 “多线程调试(/MTd)” 改为“多线程调试DLL (/MDd)”

8.构建字符串时有时需要将“1，2，3……23，24，25…….”显示成为“001,002,003…..023,024”，可使用’%3d’，如：

<code class="hljs perl has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;">spintf(neName,<span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">"person_<span class="hljs-variable" style="color: rgb(102, 0, 102); box-sizing: border-box;">%3d</span>.bmp"</span>,i);</code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul>

9.使用OpenCv里面的机器学习算法时，要保证给的labelData的数据格式为”有符号的整型数”，”CV_32FC1”就是错误的一种形式，应该使用”CV_32SC1”

顶

0

踩