blocks|key|2058171|text|如果您的相机是静止的，下面的场景是可行的：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2058172|使用背景减法方法将背景图像与前景斑点分离。|ordered-list-item|offset|length|2058173|使用形态运算改进前景结果。|2058174|检测汽车上的斑点，并消除其他斑点。|2058175|跟踪视频中的前景斑点，即二进制跟踪(只需使用这，甚至应用KF)。|2058176|entityMap|0|LINK|mutability|MUTABLE|url|https://docs.opencv.org/master/d1/dc5/tutorial_background_subtraction.html|1|https://docs.opencv.org/3.4/d3/dbe/tutorial_opening_closing_hats.html|2|https://stackoverflow.com/a/12330817/4003077|3|https://docs.opencv.org/master/dd/d6a/classcv_1_1KalmanFilter.html^0|0|2|4|0|0|2|4|1|0|0|M|1|2|S|2|3|0^^$0|@$1|2|3|4|5|6|7|10|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|11|8|@]|9|@$E|12|F|13|1|14]]|A|$]]|$1|G|3|H|5|D|7|15|8|@]|9|@$E|16|F|17|1|18]]|A|$]]|$1|I|3|J|5|D|7|19|8|@]|9|@]|A|$]]|$1|K|3|L|5|D|7|1A|8|@]|9|@$E|1B|F|1C|1|1D]|$E|1E|F|1F|1|1G]]|A|$]]|$1|M|3|-4|5|6|7|1H|8|@]|9|@]|A|$]]]|N|$O|$5|P|Q|R|A|$S|T]]|U|$5|P|Q|R|A|$S|V]]|W|$5|P|Q|R|A|$S|X]]|Y|$5|P|Q|R|A|$S|Z]]]]

If your camera is stationary the following scenario is feasible:

<ol>
<li>use ‌<a href="https://docs.opencv.org/master/d1/dc5/tutorial_background_subtraction.html" rel="nofollow noreferrer">background subtraction</a> methods to separate background image from foreground blobs.</li>
<li>Improve the foreground results using <a href="https://docs.opencv.org/3.4/d3/dbe/tutorial_opening_closing_hats.html" rel="nofollow noreferrer">morphological operations</a>.</li>
<li>Detect car blobs and remove other blobs.</li>
<li>Track foreground blobs in video i.e. binary track (simply use <a href="https://stackoverflow.com/a/12330817/4003077">this</a> or even apply <a href="https://docs.opencv.org/master/dd/d6a/classcv_1_1KalmanFilter.html" rel="nofollow noreferrer">KF</a>). </li>
</ol>

blocks|key|2058226|text|在这种情况下，一种非常基本但有效的方法可能是跟踪包围框的中心坐标，如果中心坐标只沿一个轴变化(对任何一个轴都有较小的公差)，则它是线性运动(而不是旋转)。如果x和y都变了，汽车就会在环形交叉路口行驶。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2058227|这只有一个弱点，它将检测对角线运动，但既然你看到的是一个中心环形交叉，这不应该是一个问题。|2058228|这也将是非常有效的记忆方面。|2058229|entityMap^0|0|0|0^^$0|@$1|2|3|4|5|6|7|H|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|I|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|J|8|@]|9|@]|A|$]]|$1|F|3|-4|5|6|7|K|8|@]|9|@]|A|$]]]|G|$]]

A very basic but effective approach in this scenario might be to track the center coordinates of the bounding box, if the center coordinates only change along one axis (with a small tolerance for either axis), its a linear motion (not a rotation). If both x and y change, the car is moving in the roundabout. 

This only has the weakness that it will detect diagonal motion, but since you are looking at a centered roundabout, that shouldn't be an issue. 

It will also be very efficient memory-wise.

blocks|key|1214681|text|您应该使用PCA方法，它可以计算检测到的对象的方向以及它面临的方向。您可以更改检测的阈值，以选择对象更像汽车(基于形状和颜色-+HSV转换，在您的情况下是红色)在您的图片。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1214682|链接到主成分分析(PCA)介绍|offset|length|1214683|entityMap|0|LINK|mutability|MUTABLE|url|https://docs.opencv.org/master/d1/dee/tutorial_introduction_to_pca.html^0|0|0|F|0|0^^$0|@$1|2|3|4|5|6|7|N|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|O|8|@]|9|@$D|P|E|Q|1|R]]|A|$]]|$1|F|3|-4|5|6|7|S|8|@]|9|@]|A|$]]]|G|$H|$5|I|J|K|A|$L|M]]]]

You should use PCA method, which can calculate the orientation of an detected object and which way it is facing. You can change the threshold of detection to select objects more like the cars (based upon shape and colour - a HSV conversion which in your case is red) in your picture. 

<a href="https://docs.opencv.org/master/d1/dee/tutorial_introduction_to_pca.html" rel="nofollow noreferrer">Link to an introduction to Principal Component Analysis (PCA)</a>

blocks|key|1214813|text|方法1：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1214814|++++++-+Detect+bounding+boxes+and+subtract+the+background+to+get+blobs+rotated+rectangles.|code-block|syntax|javascript|1214815|方法2：|1214816|++++++-+implement+your+own+version+of+detector+with++rotated+boxes.|1214817|方法3：|1214818|++++++-+Use+segmentation+instead++...+Unet+for+example.+|1214819|entityMap^0|0|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|Q|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|R|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|S|8|@]|9|@]|A|$]]|$1|I|3|J|5|D|7|T|8|@]|9|@]|A|$E|F]]|$1|K|3|L|5|6|7|U|8|@]|9|@]|A|$]]|$1|M|3|N|5|D|7|V|8|@]|9|@]|A|$E|F]]|$1|O|3|-4|5|6|7|W|8|@]|9|@]|A|$]]]|P|$]]

Method 1 : 

<pre><code> - Detect bounding boxes and subtract the background to get blobs rotated rectangles.
</code></pre>

Method 2 : 

<pre><code> - implement your own version of detector with rotated boxes.
</code></pre>

Method 3 : 

<pre><code> - Use segmentation instead ... Unet for example. 
</code></pre>

blocks|key|1847794|text|除了图书馆里找到的那些追踪器之外，没有别的追踪器了。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1847795|你最好的选择是过滤图像并使用查找轮廓。光流和背景减法将有助于这一点。你可以结合光流和你的汽车探测器，以排除假阳性。flow.html+subtraction.html|offset|length|1847796|entityMap|0|LINK|mutability|MUTABLE|url|https://docs.opencv.org/3.4/d4/dee/tutorial_optical_flow.html|1|https://docs.opencv.org/3.4/d1/dc5/tutorial_background_subtraction.html^0|0|1L|9|0|1V|G|1|0^^$0|@$1|2|3|4|5|6|7|P|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|Q|8|@]|9|@$D|R|E|S|1|T]|$D|U|E|V|1|W]]|A|$]]|$1|F|3|-4|5|6|7|X|8|@]|9|@]|A|$]]]|G|$H|$5|I|J|K|A|$L|M]]|N|$5|I|J|K|A|$L|O]]]]

There are no other trackers than the ones found in the library. 
Your best bet is to filter the image and use findcontours.
Optical flow and background subtraction will help with this. You can combine optical flow with your car detector to rule out false positives.
<a href="https://docs.opencv.org/3.4/d4/dee/tutorial_optical_flow.html" rel="nofollow noreferrer">https://docs.opencv.org/3.4/d4/dee/tutorial_optical_flow.html</a>
<a href="https://docs.opencv.org/3.4/d1/dc5/tutorial_background_subtraction.html" rel="nofollow noreferrer">https://docs.opencv.org/3.4/d1/dc5/tutorial_background_subtraction.html</a>

I need to track cars on the road from top-view video.

My application contain two main parts:

<ol>
<li>Detecting cars on the frame (Tensorflow trained network)</li>
<li>Tracking detected cars (opencv trackers)</li>
</ol>

I have troubles with opencv trackers. Initially i tried to different trackers, but only MOSSE is fast enough. This tracker works almost perfect for case with straight road, but i faced problems with rotating cars. This situation appears on crossroads.

As i understood, bounding box of rotated object is bigger that bbox of horizontal or vertical object. As result bbox contains big part of static background and the tracker lose target object.

Are there any alternative trackers which can track contours (not bounding boxes)?
Can i adjust quality of existing opencv trackers results by any settings or by adjusting picture? 

Schema:
<a href="https://i.stack.imgur.com/cjIjf.jpg" rel="noreferrer"><img src="https://i.stack.imgur.com/cjIjf.jpg" alt="Example"></a>

Real image:
<a href="https://i.stack.imgur.com/SSL9r.jpg" rel="noreferrer"><img src="https://i.stack.imgur.com/SSL9r.jpg" alt="Real image"></a>

Tracking of rotating objects 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

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

我需要从俯视视频中追踪道路上的车辆。我的应用程序包含两个主要部分：在车架上检测车辆(Tensorflow训练网络)跟踪检测到的汽车(opencv跟踪器)我在opencv追踪器上有麻烦。一开始我尝试不同的追踪器，但只有MOSSE足够快。这个跟踪器工作几乎完美的情况下，与直线道路，但我面临的问题，旋转汽车。这种情况出现在十字路口。据我所知，旋转物体的边框比水平或垂直物体的bbox大。因此，bbox包含

问利用opencv跟踪旋转物体
EN

回答 5

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

问利用opencv跟踪旋转物体EN

回答 5

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

问利用opencv跟踪旋转物体
EN