基于双树复小波变换的激光亮度自适应调节

Adaptive intensity of laser control based on dual-tree complex wavelet transform

在线结构光视觉测量系统中,被测表面如果同时有强反射区域和吸光率高的黑色区域时,固定不变的激光亮度投射到被测对象上的散射效果不同。在进行图像处理特征光条中心提取时带来很大困难,从而造成测量误差。为此,设计了DT-CWT激光亮度自适应调节系统。首先,通过实验分析了各档级的电压与光条图像双树复小波分解后高低频能量比的关系。然后,选择光条图像双树复小波分解后6个高低频能量比作为光条图像质量评价函数。最后,自适应地找到适合硬盘光条图像处理的合适的激光亮度。在对硬盘的测量中,光条中心提取结果恰好在能量比曲线变换比较平缓的3.5V至3.9V段。该方法满足系统测量要求,显著提高了特征光条中心提取精度。

In line structured light vision measurement system,if there are strong reflection area and the black region of high absorbance at the same time on the measured surface,scattering effect that fixed laser brightness onto the different measured objects is diverse. In the image processing and extraction of laser stripe center with great difficulty,consequently it caused error of measurement. To do this,this paper designed the adaptive intensity of laser control system based on dual-tree complex wavelet transform. First,we analyse relationship between the individual voltage levels and energy ratio of high and low frequency after dual-tree complex wavelet transform by experimentation. Then,we select six energy ratios of high and low frequency of laser tripe image by dual-tree complex wavelet decomposition as evaluation function of Image quality. In the end,it adaptive to find image of the hard disk processing of the light stripe is suitable laser intensity. In detection of the hard disk,the result of Light strip center extraction is exactly in the place of gentle energy ratio transform curve between 3. 5V to 3. 9V. The method can satisfy the measure requirement of system,and it significantly improved the extraction accuracy of laser stripe center.