Contours.md

# 寻找图像的轮廓

轮廓是由连续的点组成的曲线。轮廓与边缘很相似，但轮廓是连续的，边缘不一定都连续。

轮廓反映了物体的基本外形，常用于形状分析和物体的检测和识别。

OpenCV 提供函数 **cv.findContours()** 对二值图像寻找轮廓，函数 **cv2.drawContours()** 绘制轮廓。

**函数说明：**

```
cv.findContours(image, mode, method[, contours[, hierarchy[, offset]]]	) → contours, hierarchy
```

我们从康熙御笔书法中提取轮廓，据说这是一种经典的书画复制技法。下图从上到下分别是：
* 康熙御笔原图
* 康熙御笔碑帖图
* 康熙御笔轮廓图

![](https://gitcode.net/csdn/skill_tree_git_md_linux/-/raw/master/data/1.OpenCV初阶/2.二值图像处理/5.轮廓/ContuorsOutput.jpg)
<br/>

下面对康熙御笔寻找轮廓实现代码正确的是？

##  答案

```
import cv2 as cv

if __name__ == '__main__':
    img = cv.imread("Contours.jpg", flags=1)
    imgGray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)

    ret, thresh = cv.threshold(imgGray, 127, 255, cv.THRESH_BINARY_INV)
    image, contours, hierarchy = cv.findContours(thresh, cv.RETR_TREE, cv.CHAIN_APPROX_SIMPLE)
    contourPic = cv.drawContours(img, contours, -1, (0, 0, 255), 2)

    cv.imshow("ContourPicture", contourPic)
    cv.waitKey(0)
```

##  选项

###  使用了灰度图而不是二值图来查找轮廓

```
import cv2 as cv

if __name__ == '__main__':
    img = cv.imread("Contours.jpg", flags=1)
    imgGray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)

    ret, thresh = cv.threshold(imgGray, 127, 255, cv.THRESH_BINARY_INV)
    image, contours, hierarchy = cv.findContours(imgGray, cv.RETR_TREE, cv.CHAIN_APPROX_SIMPLE)
    contourPic = cv.drawContours(img, contours, -1, (0, 0, 255), 2)

    cv.imshow("ContourPicture", contourPic)
    cv.waitKey(0)
```

###  阈值处理时的阈值设置过大

```
import cv2 as cv

if __name__ == '__main__':
    img = cv.imread("Contours.jpg", flags=1)
    imgGray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)

    ret, thresh = cv.threshold(imgGray, 205, 255, cv.THRESH_BINARY_INV)
    image, contours, hierarchy = cv.findContours(thresh, cv.RETR_TREE, cv.CHAIN_APPROX_SIMPLE)
    contourPic = cv.drawContours(img, contours, -1, (0, 0, 255), 2)

    cv.imshow("ContourPicture", contourPic)
    cv.waitKey(0)
```

###  函数 findContours 缺少输出参数 image

```
import cv2 as cv

if __name__ == '__main__':
    img = cv.imread("Contours.jpg", flags=1)
    imgGray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)

    ret, thresh = cv.threshold(imgGray, 127, 255, cv.THRESH_BINARY_INV)
    contours, hierarchy = cv.findContours(thresh, cv.RETR_TREE, cv.CHAIN_APPROX_SIMPLE)
    contourPic = cv.drawContours(img, contours, -1, (0, 0, 255), 2)

    cv.imshow("ContourPicture", contourPic)
    cv.waitKey(0)
```