关于古建筑图像中破损点优化提取仿真

Simulation and Optimization of Broken Points in Ancient Architecture Images

对古建筑图像中对破损点的优化提取,对古建筑后续原貌恢复具有重要意义。对图像破损点的提取,需要获得破损点分量特征信息,计算图像的破损点超像素级视觉特征,完成图像中破损点优化提取。传统方法保留破损图像关键区域特征点,删除古建筑图像背景区域特征点,但忽略了计算图像的破损点超像素级视觉特征,导致提取精度偏低。提出基于小波阈值自适应修正的古建筑图像破损点提取方法。基于RAC约束的两步法来标定系统,对图像进行灰度化转换,进行图像的边缘检测和小波降噪处理,对小波降噪提纯后的古建筑图像进行破损点深度超像素特征分割,获得破损点分量特征信息,计算图像的破损点向量量化区域的超像素级视觉特征,实现破损图像破损点提取优化。仿真证明,所提方法从视觉对比和量化分析的角度实现了破损区域的轮廓分割与检测,破损点提取效果优越。

An extraction method for image breakage point of ancient architecture is proposed based on adaptive correction of wavelet threshold. Firstly, the system based on two-step method of RAC constraint is calibrated and im- age gray transformation is carried out. Then, image edge detection and wavelet denoising is conducted, and the super pixel feature of breakage point depth of ancient architectural buildings after wavelet denoising is segmented to obtain feature information of breakage point. Finally, the super pixel visual feature of quantization region of breakage point vector of image is calculated and the extraction optimization of breakage point for breakage image is realized. Simula- tion results show that the mentioned method achieves contour segmentation and detection of breakage area from visual contrast and quantitative analysis, and has good breakage point extraction performance.