工作学习第二--试错总结与学习进步

1. 图像配准的方法有很多种，此次遇到的有：https://blog.youkuaiyun.com/yuanlulu/article/details/82222119 基于维数3X3的单应性矩阵（Homography ）---基于特征的

https://blog.youkuaiyun.com/LuohenYJ/article/details/89351966（有参考代码）

 

边缘检测+角点特征+surf特征提取：https://blog.youkuaiyun.com/qq_37871215/article/details/68943628

2. 腐蚀（变瘦）与膨胀（变胖）:https://www.jianshu.com/p/05ef50ac89ac

开运算：先腐蚀后膨胀                     闭运算：先膨胀后腐蚀

kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5, 5))  # 定义结构元素
cv2.morphologyEx(img, cv2.MORPH_OPEN/cv2.MORPH_CLOSE, kernel)  

3. 读取文件夹下的图片将之转换成list：https://www.cnblogs.com/zangyu/p/5764905.html

4. 程序报错：

FindContours support only 8uC1 and 32sC1 images in function cvStartFindContours

原因：The documention of findContours is clearly saying that it can afford to take single channel images as inputs(i.e 8uc1 and 32sc1) But you are sending 3 channel image.

cv2.error: D:\Build\OpenCV\opencv-3.3.0\modules\video\src\ecc.cpp:540: error: (-7) The algorithm stopped before its convergence. The correlation is going to be minimized. Images may be uncorrelated or non-overlapped in function cv::findTransformECC

 原因：https://stackoverflow.com/questions/45997891/cv2-motion-euclidean-for-the-warp-mode-in-ecc-image-alignment-method

错误：https://stackoverflow.com/questions/50631195/getting-an-error-opencv3-4-1-c-projects-opencv-python-opencv-modules-imgproc（OpenCV（3.4.1）C：\ projects \ opencv-python \ opencv \ modules \ imgproc \ src \ thresh.cpp：1406：错误：（-215）src.type（）==（（（0）＆（（ 1 << 3） - 1））+（（（1）-1）<< 3））在函数cv :: threshold中）

5.  画矩形框：https://blog.youkuaiyun.com/wsp_1138886114/article/details/82945328

6. 获取文件夹下所有文件名：https://blog.youkuaiyun.com/u012871045/article/details/78878488

https://www.cnblogs.com/denny402/p/5123772.html

https://blog.youkuaiyun.com/qq_27261889/article/details/81151207（python+os根据文件名自动生成文本）

https://blog.youkuaiyun.com/wsLJQian/article/details/81210867（获得txt每一行）

https://blog.youkuaiyun.com/heishanlaoniu/article/details/81412314（删掉字符串中指定字符）

https://blog.youkuaiyun.com/weixin_39416561/article/details/84190234（批量修改文件名）

7. 正则表达式（还是十窍懂了九窍）

https://www.runoob.com/python/python-reg-expressions.html

https://blog.youkuaiyun.com/kpchen_0508/article/details/40921457

https://www.cnblogs.com/chuxiuhong/p/5885073.html

8：改变图片大小（PIL和cv2）

https://blog.youkuaiyun.com/LemonTree_Summer/article/details/81060829

我们习惯的坐标表示为(x,y)，在OpenCV中对图像处理时的坐标表示为(y,x)，即img.shape返回的是( 高度，宽度 )=(y,x)。然而cv2.resize(),cv2.warpAffine()函数的dst和dsize参数输入是 （宽度，高度），与img.shape相反。这里一定要注意！！很容易出错