与移动机器视觉和 OpenCV 等工具一起使用时，哪种条形码类型最可靠？
我想使用条形码来帮助机器人在家中导航。我正在研究 QR 码和其他一些类型，但我对条形码识别了解不够，无法预先了解从不同角度观看时甚至在机器人处于工作状态时都能给我带来最高成功率的变体运动。有谁知道哪种条形码类型在此类应用中表现出色？我很可能将 OpenCV 与开源条形码库之一结合使用来完成我的工作。我的用户将在打印机上打印条形码，并将其安装在坚硬的表面上，以创建放置在房屋周围战略位置的导航标牌。
I want to use bar codes to help a robot navigate a home. I was looking at QR codes and a few other kinds, but I don't know enough about bar code recognition to know up front the variant that will give me the highest success rate when viewed at different angles and perhaps even while the robot is in motion. Does anyone know which bar code type excels in applications like this one? I most likely will be using OpenCV in conjunction with one of the open source bar code libraries to do my work. My users will print the bar codes on their printer an mount them on a stiff surface to create navigation placards placed at strategic locations around the house.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（4）
QR Code 和 Data Matrix 都是不错的选择。 QRCode 更常用于最终用户应用程序，而 Data Matrix 更常见于机器人和工业应用程序。如果您需要最大的灵活性，我会推荐 Data Matrix，并且截至几年前，它仍然是稍微更健壮的代码。但由于 QR 码现在非常流行，而且有很多可用的实现，您不妨使用 QR 码。
QR 码和 Data Matrix 都是稳健的代码。它们可以承受一定程度的损坏，即使以一定角度、在不均匀的照明下查看代码，或者即使代码位于曲面上，良好的解码器也能正常工作。具有内置纠错功能的代码的优点之一是它们只有两种读取结果：成功产生正确的数据，或解码失败且没有数据。你不会被误读。
开源实现可能适用于高对比度代码（例如白纸上的黑色墨水），但如果您最终遇到问题，您可能会考虑从工业机器视觉公司购买廉价的商业智能相机。几年前，我用笔刀在一大块浅色橡胶上刻了一个数据矩阵，只要一瞥，工业质量的解码器就可以读取代码。
Both QR Code and Data Matrix are good choices. QRCode is more commonly used for end user applications, and Data Matrix is seen more often in robotic and industrial applications. If you needed maximum flexibility I'd recommend Data Matrix, and as of a few years ago it was still a marginally more robust code. But since QR Code is so popular these days and since there are so many implementations available, you might as well use QR Code.
Both QR Code and Data Matrix are robust codes. They can survive a certain amount of damage, and good decoders will work even if the code is viewed at an angle, in uneven lighting, or even if the code is located on a curved surface. One advantage of codes with built-in error correction is that they have only two read results: success that yields the correct data, or a decode failure and no data. You won't get misreads.
Open source implementations will probably work fine for high contrast codes (e.g. black ink on white paper), but if you end up having problems you might consider buying a cheap commercial smart camera from an industrial machine vision company. Some years ago I carved a Data Matrix into a hunk of light-color rubber with a pen knife, and with glancing light the code was readable to an industrial-quality decoder.
我将继续建议 QR 码
请记住，模糊必须尽可能少，并且图像必须清晰。我建议您放大 QR 码，这样分辨率就不会成为问题（也许是 50 厘米 x 50 厘米？）并且不包含大量数据。
I will go ahead and suggest QR Code
Keep in mind there must be as little blur as possible and the image must be clear. I suggest you scale the QR Codes up so resolution doesn't become a problem(50cm x 50cm maybe?) and do not include a lot of data.
zbar: http://zbar.sourceforge.net/download.html
zxing: http://code.google.com/p/zxing/
我在 ios 上使用 zbar 来构建一个非常简单的应用程序，效果很好。
zbar: http://zbar.sourceforge.net/download.html
zxing: http://code.google.com/p/zxing/
I used zbar on ios for a very simple app, it worked pretty well.
您是否查看过现有的 ar_pose 库（它是 ROS 包）或 < 使用的底层库代码>ar_pose，ARToolKit？
Have you looked at the existing ar_pose library (it's a ROS package) or the underlying library used by
ar_pose, ARToolKit?