摘要
为了提高艺术体操动作图像质量,提出基于改进Hough变换的艺术体操动作图像视觉误差校正方法。通过预先设定全局阈值与容忍度,改进Hough变换算法,利用图像灰度分布密度函数,完成动作图像空域变换,根据元素灰度值序列,滤波处理动作图像像素点,依据获取的最优分割阈值,划分图像前景与背景,采用改进的Hough变换算法,提取全部图像特征点,基于图像畸变度与镜头、轴线距离关系,结合动作图像分形码,得到分形区域偏差的最佳估算值,实现视觉误差校正。仿真实验结果表明方法校正优势显著,具有一定的有效性与可行性。
In order to improve the image quality of rhythmic gymnastics,a method of visual error correction based on improved Hough transform is proposed.By pre-setting the global threshold and tolerance,the Hough transform algorithm is improved,and the image gray distribution density function is used to complete the spatial domain transformation of the action image.According to the element gray value sequence,the action image pixels are filtered and the optimal segmentation threshold is obtained.Divide the image foreground and background,use an improved Hough transform algorithm to extract all image feature points,based on the relationship between the image distortion and the distance between the lens and the axis,combined with the action image fractal code,obtain the best estimate of the fractal area deviation,and realize the visual error correction.The simulation experiment results show that the method has significant correction advantages,and has certain validity and feasibility.
作者
于艳
YU Yan(Xianyang Normal University,Xianyang 712000 China)
出处
《自动化技术与应用》
2023年第4期45-48,共4页
Techniques of Automation and Applications
