编号和文本检测可能的OpenCV的机器人，而视频拍摄？ [英] Number and text detection possible with opencv android while video capture?
问题描述
时有可能发现同时使用OpenCV的或任何其他最佳的图像和放捕获的Android视频数字或文字;视频处理的API
Is it possible to detect numbers or text while capturing video in android using opencv or any other best image& video processing api's
推荐答案
您将需要一个有能力的OCR引擎将尝试通过任意数量的方式来检测字符。 的tesseract 是一个很好的开源引擎，它会尝试解析你的图像通过屏蔽字符。
You will need a capable OCR engine which will try to detect characters by any number of means. Tesseract is a good open source engine which will try to 'parse' your image for characters by masking.
不过，也有几个步骤或方法，你需要你给你的OCR引擎（的tesseract）图像前服用。为了确保更准确的结果，需要干净的使用二值化一些其他的常规方法，如'精明'的边缘检测沿着你的形象。这是OpenCV的帮助。
However, there are several steps or approaches you need to take before you feed your OCR engine(Tesseract) your image. In order to ensure more accurate results, you need to 'clean' your image using binarization along with a number of other conventional methods such as 'canny' edge detection. This is where 'OpenCV' can help.
此外，你应该隔离/图像中检测字符。这可以通过强大的算法来完成，如笔划宽度变换
Also, you should isolate/detect characters in images. This can be done with powerful algorithms such as the Stroke Width Transform
关于检测在Android的视频流，可以通过清洗和OCR引擎运行捕获的帧，因为他们收到过：
Regarding detection on a video stream in android, you can run your captured frames through the cleansing and OCR engine as they are recieved through:
onPreviewFrame(byte[] data, Camera camera)
此外，检查出这个应用，它允许在OCR的不断preVIEW'模式。
Also, check out this app which allows OCR in 'continuous preview' mode.
希望这有助于
这篇关于编号和文本检测可能的OpenCV的机器人，而视频拍摄？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！
