面向巡线无人机高压线实时检测与识别算法

Research on Real-time Detection and Recognition Algorithm of High-voltage Transmission Line for Inspection with Unmanned Aerial Vehicle

以巡线无人机巡航中识别高压输电线为背景,提出一种准确、实时的高压输电线检测与识别算法.首先,针对高压输电线成像是线状结构和低灰度值的特征而且其空间分布近似水平,提出一种基于方向约束的多尺度线状目标强化算法.此方法把近似水平方向的高压输电线目标强化出来的同时,能够很好地抑制竖直方向线状干扰物体和非线状背景及噪声.然后,对强化后的结果进行基于角度约束的Radon变换.由于高压输电线邻近区域的灰度分布近似,在Radon变换中引入用于识别高压输电线的识别因子,以获得高压输电线的识别结果,并抑制近似水平的干扰物体.实际的飞行试验结果表明,该算法对高压输电线识别具有很好的抗噪性、抗干扰性和实时性.

In this paper, we present an accurate, real-time high-voltage transmission line detection and recognition algorithm for the inspection of high-voltage transmission line with Unmanned Aerial Vehicle （UAV ）. Firstly according to linear structure feature of high-voltage transmission line in the horizontal direction and its low intensity in an image, we design an algorithm of automatic high- voltage transmission line detection based on multi-scale linear object enhancement with direction constraint. The algorithm can en- hance high-voltage line in the similar horizontal direction in a frame image while restraining linear objects in the vertical direction and other nonlinear objects and noise. Then, Radon transform based on the angle constraint is applied to recognize the high-voltage line within a frame image with image segmentation. Because the intensity distribution near high voltage transmission lines are similar, we can introduce the recognition factor for the high voltage transmission lines into the Radon Transform in order to get the recognition resuits and repress the disturbances from horizontal objects. The actual flight tests have been completed. The results of our experiment show that our algorithm has good performance in anti-noise, robustness against distortion and real-time.