entityMap|0|type|IMAGE|mutability|IMMUTABLE|data|imageUrl|https://ask.qcloudimg.com/http-save/yehe-900000/c7d483cf3a7782dc5a08b55e51bff9dd.png|imageAlt|1|https://ask.qcloudimg.com/http-save/yehe-900000/3f52947ad4d95f7d0a384e77504ebfc8.png|2|https://ask.qcloudimg.com/http-save/yehe-900000/539e51d7a610b839e2341032f59c81f1.png|3|https://ask.qcloudimg.com/http-save/yehe-900000/a66b23c07d7a86727091b5a52ac8a9fa.png|blocks|key|b3fao|text|你可以取第一个图像数组的阈值。它看起来是这样的：|unstyled|depth|inlineStyleRanges|entityRanges|1qr32|​|14ef7|?|atomic|offset|length|2sg34|3b4pu|在这里，可以清楚地看到墨水标记和字母的像素值的差异。阈值之后，它看起来是这样的：|2d3ao|fd2gj|2kflh|4tm6q|现在可以通过关闭来删除墨迹。然而，它也会减小字母的大小。因此，在逐位OR之后执行侵蚀，以获得没有墨迹标记的掩码。|dl4ea|dbe8p|8pkir|3u4si|但是，如果您希望字母看起来像原始图像，您可以将掩码存储在255s的数值数组中，并对原始图像执行逐位“或”运算。|e27uo|7i2bv|vl99|8vlkj|我使用的完整代码是：|2859b|img+=+cv2.imread('ink_mark.png')
wimg+=+img[:,+:,+0]
ret,thresh+=+cv2.threshold(wimg,100,255,cv2.THRESH_BINARY)

kernel+=+np.ones((7,+7),+np.uint8)
closing+=+cv2.morphologyEx(thresh,+cv2.MORPH_CLOSE,+kernel)
erosion+=+cv2.erode(closing,+kernel,+iterations+=+1)
mask+=+cv2.bitwise_or(erosion,+thresh)
white+=+np.ones(img.shape,np.uint8)*255
white[:,+:,+0]+=+mask
white[:,+:,+1]+=+mask
white[:,+:,+2]+=+mask
result+=+cv2.bitwise_or(img,+white)

cv2.imshow('image',+result)
cv2.waitKey(0)
cv2.destroyAllWindows()|code-block|syntax|javascript|4n9bj^0|0|0|0|1|0|0|0|0|0|0|1|1|0|0|0|0|0|1|2|0|0|0|0|0|1|3|0|0|0|0^^$0|$1|$2|3|4|5|6|$7|8|9|-4]]|A|$2|3|4|5|6|$7|B|9|-4]]|C|$2|3|4|5|6|$7|D|9|-4]]|E|$2|3|4|5|6|$7|F|9|-4]]]|G|@$H|I|J|K|2|L|M|1K|N|@]|O|@]|6|$]]|$H|P|J|Q|2|L|M|1L|N|@]|O|@]|6|$]]|$H|R|J|S|2|T|M|1M|N|@]|O|@$U|1N|V|1O|H|1P]]|6|$]]|$H|W|J|Q|2|L|M|1Q|N|@]|O|@]|6|$]]|$H|X|J|Y|2|L|M|1R|N|@]|O|@]|6|$]]|$H|Z|J|Q|2|L|M|1S|N|@]|O|@]|6|$]]|$H|10|J|S|2|T|M|1T|N|@]|O|@$U|1U|V|1V|H|1W]]|6|$]]|$H|11|J|Q|2|L|M|1X|N|@]|O|@]|6|$]]|$H|12|J|13|2|L|M|1Y|N|@]|O|@]|6|$]]|$H|14|J|Q|2|L|M|1Z|N|@]|O|@]|6|$]]|$H|15|J|S|2|T|M|20|N|@]|O|@$U|21|V|22|H|23]]|6|$]]|$H|16|J|Q|2|L|M|24|N|@]|O|@]|6|$]]|$H|17|J|18|2|L|M|25|N|@]|O|@]|6|$]]|$H|19|J|Q|2|L|M|26|N|@]|O|@]|6|$]]|$H|1A|J|S|2|T|M|27|N|@]|O|@$U|28|V|29|H|2A]]|6|$]]|$H|1B|J|Q|2|L|M|2B|N|@]|O|@]|6|$]]|$H|1C|J|1D|2|L|M|2C|N|@]|O|@]|6|$]]|$H|1E|J|1F|2|1G|M|2D|N|@]|O|@]|6|$1H|1I]]|$H|1J|J|-4|2|L|M|2E|N|@]|O|@]|6|$]]]]

You can take threshold of the first array of image. It looks like this:

<a href="https://i.stack.imgur.com/J2cXS.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/J2cXS.png" alt="first array"></a>

Here it is clearly visible the difference in pixel values of the ink mark and the letters. After thresholding it looks like:

<a href="https://i.stack.imgur.com/pqwlP.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/pqwlP.png" alt="thresholded image"></a>

The ink mark can now be removed via closing. However it will reduce the size of letters as well. Therefore erosion is performed followed by a bitwise OR to obtain our mask without the ink mark. 

<a href="https://i.stack.imgur.com/4EnUI.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/4EnUI.png" alt="mask"></a>

If however you want the letters to look like the original image you can store the mask in a numpy array of 255s and perform it bitwise OR with original image. 

<a href="https://i.stack.imgur.com/AJAbQ.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/AJAbQ.png" alt="result"></a>

The full code I have used is:

<pre><code>img = cv2.imread('ink_mark.png')
wimg = img[:, :, 0]
ret,thresh = cv2.threshold(wimg,100,255,cv2.THRESH_BINARY)

kernel = np.ones((7, 7), np.uint8)
closing = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
erosion = cv2.erode(closing, kernel, iterations = 1)
mask = cv2.bitwise_or(erosion, thresh)
white = np.ones(img.shape,np.uint8)*255
white[:, :, 0] = mask
white[:, :, 1] = mask
white[:, :, 2] = mask
result = cv2.bitwise_or(img, white)

cv2.imshow('image', result)
cv2.waitKey(0)
cv2.destroyAllWindows()
</code></pre>

entityMap|0|type|IMAGE|mutability|IMMUTABLE|data|imageUrl|https://ask.qcloudimg.com/http-save/yehe-900000/de79349d59a26c222685b5c6df4839ef.png|imageAlt|1|https://ask.qcloudimg.com/http-save/yehe-900000/81b6bd88621e4c1bc9ad2249d3f88fbc.png|blocks|key|1mcgp|text|试着使用inpaint。首先创建墨迹的蒙版：|unstyled|depth|inlineStyleRanges|entityRanges|f1f1m|hsv+=+cv2.cvtColor(img,+cv2.COLOR_BGR2HSV)+
lower_blue+=+np.array([100,50,50])+
upper_blue+=+np.array([150,255,255])+
kernel+=+np.ones((5,5),np.uint8)
mask+=+cv2.inRange(hsv,+lower_blue,+upper_blue)
mask+=+cv2.dilate(mask,kernel,iterations+=+4)|code-block|syntax|javascript|5svpu|​|4kvil|?|atomic|offset|length|b3psl|5g0r9|使用inpaint函数在遮罩为白色的区域进行绘制。OpenCV将丢弃原始像素，并使用猜测哪些像素应该放在那里。|145fv|dst+=+cv2.inpaint(img,+mask,+3,+cv2.INPAINT_TELEA)|1db00|707ts|e34sf|3bp4d^0|0|0|0|0|1|0|0|0|0|0|0|0|1|1|0|0^^$0|$1|$2|3|4|5|6|$7|8|9|-4]]|A|$2|3|4|5|6|$7|B|9|-4]]]|C|@$D|E|F|G|2|H|I|16|J|@]|K|@]|6|$]]|$D|L|F|M|2|N|I|17|J|@]|K|@]|6|$O|P]]|$D|Q|F|R|2|H|I|18|J|@]|K|@]|6|$]]|$D|S|F|T|2|U|I|19|J|@]|K|@$V|1A|W|1B|D|1C]]|6|$]]|$D|X|F|R|2|H|I|1D|J|@]|K|@]|6|$]]|$D|Y|F|Z|2|H|I|1E|J|@]|K|@]|6|$]]|$D|10|F|11|2|N|I|1F|J|@]|K|@]|6|$O|P]]|$D|12|F|R|2|H|I|1G|J|@]|K|@]|6|$]]|$D|13|F|T|2|U|I|1H|J|@]|K|@$V|1I|W|1J|D|1K]]|6|$]]|$D|14|F|R|2|H|I|1L|J|@]|K|@]|6|$]]|$D|15|F|-4|2|H|I|1M|J|@]|K|@]|6|$]]]]

Try using inpaint. First create a mask of the ink: 

<pre><code>hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV) 
lower_blue = np.array([100,50,50]) 
upper_blue = np.array([150,255,255]) 
kernel = np.ones((5,5),np.uint8)
mask = cv2.inRange(hsv, lower_blue, upper_blue)
mask = cv2.dilate(mask,kernel,iterations = 4)
</code></pre>

<a href="https://i.stack.imgur.com/PfK6f.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/PfK6f.png" alt="mask"></a>

Use the inpaint function to paint in areas where the mask it white. OpenCV will throw away the original pixels, and use guess which pixels should go there.

<pre><code>dst = cv2.inpaint(img, mask, 3, cv2.INPAINT_TELEA)
</code></pre>

<a href="https://i.stack.imgur.com/AJaw1.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/AJaw1.png" alt="enter image description here"></a>

I have an image like this,

<a href="https://i.stack.imgur.com/DPqal.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/DPqal.png" alt="enter image description here"></a>

As you can see, there is a pen mark in the image. I want to remove that mark. How to do it in OpenCV.? 

I tried converting it to HSV, creating a mask with blue range and removing it using the code.

<pre><code>hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV) 
lower_blue = np.array([110,50,50]) 
upper_blue = np.array([130,255,255]) 

mask = cv2.inRange(hsv, lower_blue , upper_blue ) 
res = cv2.bitwise_and(img, img, mask= mask) 
</code></pre>

It is not working as needed. All the text gets removed. How to fix this.?

Remove ink mark from paper using OpenCV

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

教程

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云智能顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

EdgeOne AI 安全实战专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋

腾讯云代码助手

云原生构建

TAPD 敏捷项目管理

Cloud Studio

SDK中心

API中心

命令行工具

涵盖代码开发、场景应用、自动测试全流程，助你从零构建专属AI助手

一站式MCP教程库，解锁AI应用新玩法

聚焦“写作效率、视觉美观与运行性能”三方面进行全面升级，为您提供更高效、稳定的创作环境

社区富文本&Markdown编辑器全新改版上线，欢迎大家体验!

诚挚邀请您参与本次调研，分享您的真实使用感受与建议。您的反馈至关重要，感谢您的支持与参与！

社区新版编辑器体验调研

 我有一个这样的图像， ​ ? ​ 如您所见，图像中有一个笔迹。我想去掉那个印记。如何在OpenCV中实现？ 我尝试将其转换为HSV，创建一个蓝色范围的蒙版，并使用代码删除它。 hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV) lower_blue = np.array([110,5...

问使用OpenCV去除纸张上的墨迹
EN

Stack Overflow用户

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问使用OpenCV去除纸张上的墨迹EN