opencv实现靶纸弹孔识别计数功能

任务是对胸环靶进行弹孔识别与计数。靶纸图片是从网上找的。
具体过程主要是先对图片进行灰度化处理,接着进行阈值分割,将弹孔与背景分割开。再腐蚀并膨胀(形态学开运算),取得较好的分割效果。最后通过轮廓提取,对弹孔进行识别与计数。

1. 靶纸图片

靶纸如下图所示:
靶纸图片

2. 灰度化处理

通过opencv的cv2.cvtColor()函数对图像进行灰度化。
函数cv2.cvtColor(input_image ,flag),flag是转换类型:
BGR和灰度图的转换使用 cv2.COLOR_BGR2GRAY

# src.jpg是图像的名称
img = cv2.imread('src.jpg')
# 灰度化函数,结果为GrayImage
GrayImage = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cv2.imshow('GrayImage', GrayImage)
cv2.waitKey(0)

灰度化结果如图所示:
灰度化结果图

3. 二值化处理

通过opencv的cv2.threshold()函数对得到的灰度图进行阈值分割,二值化处理。
函数参数参考参数描述

#
ret, thresh = cv2.threshold(GrayImage, 70,255,cv2.THRESH_BINARY_INV)

二值化结果图如图所示:
可以看出,阈值化处理的结果还有一些背景没有去除。
二值化结果图

3. 形态学处理(开运算)

# 开运算处理,结果图为opening
kernel = np.ones((3, 3), np.uint8)
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)

开运算结果如图所示:
开运算结果图

4. 轮廓提取,画轮廓

通过opencv的cv2.findContours()函数对开运算处理后的二值化图像进行轮廓提取,cv2.rectangle()函数。

# 轮廓提取函数
contours, hierarchy = cv2.findContours(opening, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
for i in range(0,len(contours)):
    x, y, w, h = cv2.boundingRect(contours[i])
    cv2.rectangle(img, (x,y), (x+w,y+h), (0,0,255), 2)

结果如图所示:
弹孔识别结果图

5. 弹孔计数

num = len(contours)
print("弹孔数量", num)

输出结果为3

6. 结果分析

从图中的结果可以看出,原靶纸图片应该为4个弹孔,但实际上只识别到3个弹孔。
分析原因:主要是在二值化步骤中,10环中心区域是白色的背景,导致10环处的弹孔较亮,大于我们阈值处理选择的灰度值70,从而被当成背景而未被识别。
处理:在二值化之前,应该先对10环内的区域进行处理,分割出内部的弹孔或使其灰度值降低。

7. 未完待续…

之后有时间再进行改进吧…

GitHub 加速计划 / opencv31 / opencv
218
20
下载
OpenCV: 开源计算机视觉库
最近提交(Master分支:29 天前 )
b6447542 Fix invalid memory access in USAC #27865 ### Pull Request Readiness Checklist Fix #27863 See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request - [x] I agree to contribute to the project under Apache 2 License. - [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV - [x] The PR is proposed to the proper branch - [x] There is a reference to the original bug report and related work - [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable Patch to opencv_extra has the same branch name. - [ ] The feature is well documented and sample code can be built with the project CMake 22 小时前
19c41c29 Improved blur #27822 * Perform row and column filter operations in a single pass. * Temporary storage of intermediate results are avoided. * Impacts 32F and 64F inputs for ksize <=5. ### Pull Request Readiness Checklist See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request - [x] I agree to contribute to the project under Apache 2 License. - [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV - [x] The PR is proposed to the proper branch - [ ] There is a reference to the original bug report and related work - [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable Patch to opencv_extra has the same branch name. - [ ] The feature is well documented and sample code can be built with the project CMake 1 天前
Logo

旨在为数千万中国开发者提供一个无缝且高效的云端环境,以支持学习、使用和贡献开源项目。

更多推荐