视觉导航智能汽车路径识别图像处理算法研究

On Image Processing Algorithm of Intelligent Car with Visual Navigation

复杂环境下图像快速准确处理是智能汽车稳定运行的前提和保证.该文针对由于光线等因素造成图像对比度不明显问题,提出了线性变换方法处理图像;针对由于图像边缘点的干扰而造成中位数处理失败,提出了中值滤波和T型窗口算法;基于Roberts算子的高精度和Sobel算子的高运算效率,提出了一种融合两种算子的融合算法用以改进边缘检测算法.实际应用表明,上述算法可以过滤边缘噪音,有效地去除干扰噪音,过滤后整幅图像的边缘更连续平滑,融合算法的计算量远远小于单一算子,算法的实时性和准确性大幅提高.

Image fast processing under complicated environment is the prerequisite and guarantee for the stable operation of smart car.In view of the image contrast is not obvious due to factors such as light,lin-ear transformation method has been proposed to deal with image.For due to the interference of image edge points median treatment failure,median filtering and T window algorithm has been proposed.Roberts op-erator of high precision and Sobel operator based high operation efficiency,a fusion of two operator fusion algorithm for edge detection algorithm has been proposed in this paper.Edge of practical application show that the algorithm can filter the noise and interference effectively remove noise,filtering after the edge of the whole image more continuous smooth and fusion algorithm is far less than the amount of calculation of a single operator,improve the real-time performance and accuracy of the algorithm.