论文部分内容阅读
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.
Based on the background of high-voltage transmission line in unmanned aerial vehicle cruise line inspection, an accurate and real-time high-voltage transmission line detection and identification algorithm is proposed.Firstly, the imaging of high voltage transmission line is characterized by linear structure and low gray value, and its Aiming at the approximate level of spatial distribution, a multi-scale linear target enhancement algorithm based on directional constraint is proposed.This method can strengthen the target of high-voltage transmission line in the horizontal direction and suppress the vertical linear interference objects and non- Linear background and noise.Then, the angle-constrained Radon transform is applied to the enhanced result.Due to the similar grayscale distribution in the vicinity of high-voltage transmission line, the recognition factor for identifying high-voltage transmission line is introduced into Radon transform to obtain High-voltage transmission line identification results, and suppress the interference objects of approximate level.The actual flight test results show that the algorithm has good noise immunity, anti-interference and real-time performance for high voltage transmission line identification.