我正在尝试使用hog描述符算法执行对象检测。我在我的Ubuntu 12.04系统中完成了它,处理时间为300毫秒。然后我尝试使用beagle board xm rev c3执行相同操作然后我遇到了一些问题。主要问题是它需要的处理时间太长,抓取图像和检测物体的时间差不多是45秒。通过将双精度更改为单精度,我可以将处理时间缩短到8.5秒。
But still it takes more than 8 seconds.As in my knowledge it is due to one single function
hog.detectMultiScale(img, found, THRESHOLD, cv::Size(2,2), cv::Size(0,0), 1.05, 2);
which takes 8 seconds to process.
I saw in some sites that this function require huge processing power to process so we need some better GPU.
But in our case beagle-board xm has 1GHz processor and 512 MB Ram.
(My Ubuntu system has 2 GHz processor and 1 GB Ram)
我正在处理320x240像素的图像。但是我通过裁剪来拍摄整个图像的一小部分。然后它在系统中执行300ms,但在beagle board中执行8.5秒。
我使用的是opencv 2.4.2。
我需要将处理时间从8秒减少到至少2秒以下。
Is any other solution?
Could anyone help me?