blocks|key|859386|text|在你的图像中，颜色似乎是一个比强度更好的分割指标。尝试将其转换为HSV，然后在H通道上运行OTSU。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|859387|巨蟒|859388|hsv+=+cv2.cvtColor(image,+cv2.cv.CV_BGR2HSV)
cv2.imshow('hsv',+hsv[:,:,0])
(thresh,+im_bw)+=+cv2.threshold(hsv[:,:,0],+128,+255,+cv2.THRESH_BINARY+%7C+cv2.THRESH_OTSU)
cv2.imshow('OTSU',+im_bw)|code-block|syntax|javascript|859389|给予(hsv)|859390|​|859391|📷|atomic|offset|length|859392|859393|然后(OTSU)|859394|859395|859396|859397|一点腐蚀和扩张，你应该是好的去|859398|entityMap|0|IMAGE|mutability|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/df9509666c17cae5d2c6a1b24f005ee1.jpg|imageAlt|1|https://developer.qcloudimg.com/http-save/yehe-900000/49602c8e604db381abaecdabbc8d64a1.jpg^0|0|0|0|0|0|0|1|0|0|0|0|0|0|1|1|0|0|0^^$0|@$1|2|3|4|5|6|7|1A|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|1B|8|@]|9|@]|A|$]]|$1|D|3|E|5|F|7|1C|8|@]|9|@]|A|$G|H]]|$1|I|3|J|5|6|7|1D|8|@]|9|@]|A|$]]|$1|K|3|L|5|6|7|1E|8|@]|9|@]|A|$]]|$1|M|3|N|5|O|7|1F|8|@]|9|@$P|1G|Q|1H|1|1I]]|A|$]]|$1|R|3|L|5|6|7|1J|8|@]|9|@]|A|$]]|$1|S|3|T|5|6|7|1K|8|@]|9|@]|A|$]]|$1|U|3|L|5|6|7|1L|8|@]|9|@]|A|$]]|$1|V|3|N|5|O|7|1M|8|@]|9|@$P|1N|Q|1O|1|1P]]|A|$]]|$1|W|3|L|5|6|7|1Q|8|@]|9|@]|A|$]]|$1|X|3|Y|5|6|7|1R|8|@]|9|@]|A|$]]|$1|Z|3|-4|5|6|7|1S|8|@]|9|@]|A|$]]]|10|$11|$5|12|13|14|A|$15|16|17|-4]]|18|$5|12|13|14|A|$15|19|17|-4]]]]

It seems like color is a much better indicator for segmentation in your image than intensity. Try converting it to HSV and then running OTSU on the H channel.

in python

<pre><code>hsv = cv2.cvtColor(image, cv2.cv.CV_BGR2HSV)
cv2.imshow('hsv', hsv[:,:,0])
(thresh, im_bw) = cv2.threshold(hsv[:,:,0], 128, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)
cv2.imshow('OTSU', im_bw)
</code></pre>

gives (hsv)

<img src="https://i.stack.imgur.com/ZTPLF.jpg" alt="enter image description here">

and then (OTSU)

<img src="https://i.stack.imgur.com/DT8Cy.jpg" alt="enter image description here">

A little eroding and dilating and you should be good to go

blocks|key|5277|text|您可能对openCV使用的openCV自适应阈值感兴趣。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|5278|我使用了自适应平均阈值。您可能需要使用一些参数，但如果您的图像是相似的(相同的大小等)，希望不会有太多的调整需要。|5279|#+Smooth+image
filtered+=+cv2.adaptiveThreshold(input_image.astype(np.uint8),+255,+cv2.ADAPTIVE_THRESH_MEAN_C,+cv2.THRESH_BINARY,+41,+3)

#+Some+morphology+to+clean+up+image
kernel+=+np.ones((5,5),+np.uint8)
opening+=+cv2.morphologyEx(filtered,+cv2.MORPH_OPEN,+kernel)
closing+=+cv2.morphologyEx(opening,+cv2.MORPH_CLOSE,+kernel)|code-block|syntax|javascript|5280|结果：|5281|​|5282|📷|atomic|offset|length|5283|5284|5285|5286|5287|5288|5289|5290|entityMap|0|IMAGE|mutability|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/d31c12d982f8568d3a4edac8353961d7.png|imageAlt|1|https://developer.qcloudimg.com/http-save/yehe-900000/95fccb77faf2cb10da1e14ed5152f392.png|2|https://developer.qcloudimg.com/http-save/yehe-900000/fa58f9bc12295189146268ea2e1dea31.png^0|0|0|0|0|0|0|1|0|0|0|0|0|1|1|0|0|0|0|1|2|0|0^^$0|@$1|2|3|4|5|6|7|1B|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|1C|8|@]|9|@]|A|$]]|$1|D|3|E|5|F|7|1D|8|@]|9|@]|A|$G|H]]|$1|I|3|J|5|6|7|1E|8|@]|9|@]|A|$]]|$1|K|3|L|5|6|7|1F|8|@]|9|@]|A|$]]|$1|M|3|N|5|O|7|1G|8|@]|9|@$P|1H|Q|1I|1|1J]]|A|$]]|$1|R|3|L|5|6|7|1K|8|@]|9|@]|A|$]]|$1|S|3|L|5|6|7|1L|8|@]|9|@]|A|$]]|$1|T|3|N|5|O|7|1M|8|@]|9|@$P|1N|Q|1O|1|1P]]|A|$]]|$1|U|3|L|5|6|7|1Q|8|@]|9|@]|A|$]]|$1|V|3|L|5|6|7|1R|8|@]|9|@]|A|$]]|$1|W|3|N|5|O|7|1S|8|@]|9|@$P|1T|Q|1U|1|1V]]|A|$]]|$1|X|3|L|5|6|7|1W|8|@]|9|@]|A|$]]|$1|Y|3|-4|5|6|7|1X|8|@]|9|@]|A|$]]]|Z|$10|$5|11|12|13|A|$14|15|16|-4]]|17|$5|11|12|13|A|$14|18|16|-4]]|19|$5|11|12|13|A|$14|1A|16|-4]]]]

You might be interested in <a href="http://docs.opencv.org/trunk/doc/py_tutorials/py_imgproc/py_thresholding/py_thresholding.html#adaptive-thresholding" rel="noreferrer">these</a> adaptive thresholds used by openCV. 

I used the adaptive mean threshold. You may have to play with the parameters a bit, but if your images are similar (same size etc), hopefully there won't be too much tweaking required.

<pre class="lang-python prettyprint-override"><code># Smooth image
filtered = cv2.adaptiveThreshold(input_image.astype(np.uint8), 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 41, 3)

# Some morphology to clean up image
kernel = np.ones((5,5), np.uint8)
opening = cv2.morphologyEx(filtered, cv2.MORPH_OPEN, kernel)
closing = cv2.morphologyEx(opening, cv2.MORPH_CLOSE, kernel)
</code></pre>

Results:

<img src="https://i.stack.imgur.com/fBvMY.png" alt="enter image description here">

<img src="https://i.stack.imgur.com/uXG4V.png" alt="enter image description here">

<img src="https://i.stack.imgur.com/80Ip2.png" alt="enter image description here">

blocks|key|1935409|text|下面的代码..。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1935410|im=cv2.adaptiveThreshold(img,255,cv2.ADAPTIVE_THRESH_MEAN_C,cv2.THRESH_BINARY,111,3)
cv2.imshow('mkm',im)`|code-block|syntax|javascript|1935411|..。取得了良好的效果：|1935412|​|1935413|📷|atomic|offset|length|1935414|1935415|entityMap|0|IMAGE|mutability|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/e7a4e540d7c5ee6bd8c0529ef77b88de.png|imageAlt^0|0|0|0|0|0|1|0|0|0^^$0|@$1|2|3|4|5|6|7|Z|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|10|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|11|8|@]|9|@]|A|$]]|$1|I|3|J|5|6|7|12|8|@]|9|@]|A|$]]|$1|K|3|L|5|M|7|13|8|@]|9|@$N|14|O|15|1|16]]|A|$]]|$1|P|3|J|5|6|7|17|8|@]|9|@]|A|$]]|$1|Q|3|-4|5|6|7|18|8|@]|9|@]|A|$]]]|R|$S|$5|T|U|V|A|$W|X|Y|-4]]]]

The following code ...

<pre><code>im=cv2.adaptiveThreshold(img,255,cv2.ADAPTIVE_THRESH_MEAN_C,cv2.THRESH_BINARY,111,3)
cv2.imshow('mkm',im)`
</code></pre>

... gives a good result:

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

blocks|key|4944|text|如果单个阈值不足以分离所有图像，则可以使用两个阈值尝试分水岭算法。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|4945|使用一个高阈值来获得一幅图像，其中一段一定是一个数字的一部分，而一个高的逆阈值则可以得到一幅绝对不是数字一部分的部分的图像。|4946|对这两幅图像进行一定程度的侵蚀，以增加确定性。|4947|然后用这2幅图像作为流域的种子。|4948|这是一个答案，这是在哪里做的|4949|entityMap|0|LINK|mutability|MUTABLE|url|http://docs.opencv.org/modules/imgproc/doc/miscellaneous_transformations.html#watershed|1|https://stackoverflow.com/a/11438165/145999^0|R|5|0|0|0|0|0|0|E|1|0^^$0|@$1|2|3|4|5|6|7|V|8|@]|9|@$A|W|B|X|1|Y]]|C|$]]|$1|D|3|E|5|6|7|Z|8|@]|9|@]|C|$]]|$1|F|3|G|5|6|7|10|8|@]|9|@]|C|$]]|$1|H|3|I|5|6|7|11|8|@]|9|@]|C|$]]|$1|J|3|K|5|6|7|12|8|@]|9|@$A|13|B|14|1|15]]|C|$]]|$1|L|3|-4|5|6|7|16|8|@]|9|@]|C|$]]]|M|$N|$5|O|P|Q|C|$R|S]]|T|$5|O|P|Q|C|$R|U]]]]

If a single threshold is not enough to separate all images, you could try the <a href="http://docs.opencv.org/modules/imgproc/doc/miscellaneous_transformations.html#watershed" rel="nofollow noreferrer">Watershed algorithm</a> with two thresholds.

Use a high threshold to get an image with segments that are definitely part of a digit, and a high inverse threshold to get an image with segments that are definitely not part of a digit.

Erode both images a bit to increase the certainty.

Then use the 2 images as seeds for Watershed.

<a href="https://stackoverflow.com/a/11438165/145999">Here is an answer where this is done</a>

I'm in search for best Adaptive Threshold method for image binarization. But I have any problems with dark and blurry image. 
Input image: 
<img src="https://i.stack.imgur.com/51H1M.png" alt="Let the Image load...">

and when I use Adaptive threshold method I receive this 
Output Image: 
<img src="https://i.stack.imgur.com/KX86P.jpg" alt="Let the Image load...">

This is not good for me! 
So, could someone help me fix this problem?

<hr>

another image : <img src="https://i.stack.imgur.com/xu3So.jpg" alt="enter image description here">

and : <img src="https://i.stack.imgur.com/V2sfG.png" alt="enter image description here">

the first seem very bad with @Hammer'solution (i must chose c channel) , the second i can use adaptive threshold normal . 
so i want to find the best solution for all cases . 

thank Again !

Adaptive threshold with blurry image

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

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

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

我在寻找图像二值化的最佳自适应阈值方法。但是我对黑暗和模糊的图像有任何问题。输入图像：​当我使用自适应阈值方法时，我会收到以下信息输出图像​这对我不好！有人能帮我解决这个问题吗？另一幅图片：​以及：​首先，使用@Hammer的解决方案(我必须选择c通道)，第二，我可以使用自适应阈值正常。所以我想为所有案件找到最好的解决方案。再次感谢！

问模糊图像的自适应阈值
EN

回答 4

Stack Overflow用户

Stack Overflow用户

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

问模糊图像的自适应阈值EN

回答 4

Stack Overflow用户

Stack Overflow用户

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

问模糊图像的自适应阈值
EN