高等级公路环境景观视觉图像增强算法仿真

为了提高公路环境景观视觉图像的清晰度、失真度,提出了基于变尺度Retinex的公路环境景观视觉图像增强算法,利用最小线性均方误差准则计算期望小波系数的部分方差估计值,获取基本频带系数的滤波器,利用上述滤波器平滑公路环境景观视觉图像噪声,对去噪后的景观视觉图像进行整体自适应伽马校正,计算尺度滤波器的高斯核函数增强景观视觉图像边缘信息,通过线性拉伸方式改善视觉图像的局部对比度,得到增强后的觉图像,完成了公路环境景观视觉图像增强。通过仿真证明,所提算法能够均衡景观视觉图像亮度,丰富图像色彩,增强视觉图像信息熵、平均梯度和均方差系数,降低图像偏差,保证视觉图像的信息量和清晰度。

Simulation of Visual Image Enhancement Algorithm for High-Grade Highway Environment Landscape

In order to improve the clarity and distortion of visual image for road environment landscape, this article puts forward an algorithm to enhance the landscape visual image in road environment based on variable-scale Retinex. The minimum linear mean square error criterion was used to calculate the partial variance estimation value of expected wavelet coefficient, so that the filter of base band coefficient was obtained. Then, the above filter was used to smooth the noise in landscape image of visual image for road environment landscape. After that, the overall adaptive Gamma correction was performed on the de-noised landscape visual image, and the Gaussian kernel function of scaling filter was calculated to enhance the edge information. Through the linear stretch, the local contrast of visual image was improved and the enhanced image was obtained. Thus, the enhancement of visual image in road environment landscape was completed. Simulation results show that the proposed algorithm can balance the brightness of landscape visual image and enrich image color. Meanwhile, this algorithm can enhance the information entropy of visual image, average gradient and mean square error coefficient, reduce the image deviation, and ensure the information volume and clarity of visual image.