自主预测？ [英] Autonomous Predictions?
问题描述
你好，
我和我的团队正在努力寻找一个我们可以训练以检测车牌的程序，然后连续自主地运行该预测。我们已经搜索了一段时间，但没有发现任何迹象表明Azure可以执行自治部分？
My team and I are trying to find a program that we can train to detect license plates, and then run that prediction continuously and autonomously. We've been searching for a while and haven't found anything to indicate that Azure can do the autonomous part?
如果我的问题有点不清楚：我们想要上传，比方说，显示牌照并且具有预测API（1）的1000个静止帧检测每个板，（2）给出置信度，以及（3）如果置信度
是，则将板的文本读出放入单独的文件中超过0.35。
In case my question's a little unclear: we want to upload, say, 1000 still frames which show a license plate and have the prediction API (1) detect each plate, (2) give a confidence, and (3) do a text readout of the plate into a separate file if the confidence is over 0.35.
我们不想选择最终无法满足我们要求的软件，我们无法弄清楚Azure是否能够做到这一点我们自己的未来。
We don't want to chose a software that ultimately can't do what we're asking, and we can't figure out if Azure can do it with our own futzing around.
任何帮助都将不胜感激！
Any help would be much appreciated!
推荐答案
您好，
您可以使用自定义视觉ai服务创建车牌模块（对象识别），然后使用OCR（文本）模块确定实际文本。
You can use the custom vision ai service to create a license plates module (object recognition) and then use the OCR (text) module to determine the actual text.
https://nicksnettravels.builttoroam.com/post/2018/05/24/Building-a-Number-Plate-Identification-Service-in-5-Minutes-with-Microsofts-Custom-Vision-Service.aspx
谢谢&问候
Ram
这篇关于自主预测？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
