为了实现阀厅智能巡检机器人系统对设备的精确测温，需要对阀厅中红外图像设备精确定位，文中提出了一种新的定位红外设备的方法。对已有的一幅红外图像，先进行图像预处理，然后使用SURF （Speeded Robust Features）算法提取整幅图像的特征，最终将这幅图像的特征点与待检测图像的特征点进行匹配，便可精确的定位出待检测的设备。实验证明，该方法对红外图像的设备定位的效果较好，精确度也较高。%In order to realize accurate temperature-measuring in the intelligent inspection robot of valve hall, we need accurate positioning for the equipment in infrared image, a new method of equipment positioning in infrared image put forward. First, making pre-processing to the standard infrared image,and then,using the method of SURF (Speeded Robust Features) to extract the features of whole standard image, finally, matching the features of standard image and the detective image and the detective equipment will be positioned accurately. Test proves that the method of equipment position in infrared image has a good effect and high precision.
展开▼
