激光气泡检测中图像的边缘正则化复原

Edge Regularization Based Image Restoration for Laser Bubble Detection

基于光学成像的激光气泡检测已被广泛应用于工业领域,如气密性检测等。然而,检测图像的降质严重影响了测量的准确性。为了提高气泡检测的图像质量,提出一种估计动态调制传递函数进行边缘正则化的半盲图像复原法,采用客观图像的质量指标来对半盲解卷积算法进行约束,并将图像复原法与二值形态滤波比较。实验结果证明了该方法的有效性,由此可以得出该方法能有效提高激光气泡检测的成像质量,保留气泡检测图像的边缘细节,以达到更高的检测精度。

The laser bubble detection based on optical photographic has been widely used in industrial fields,such as air tightness measurement. However,the accuracy of the detection is seriously affected by the degradation of detecting images. In order to improve the visual quality of bubble detection,an image semi-blind restoration method with edge regularization is proposed,along with the estimation of dynamic modulation transfer function. Objective image quality metrics are used to limit the deconvolution iterations of semi-blind deconvolution algorithm. Performance of the binary morphology filter is compared and improvement by the proposed method can be seen from experimental results.It can be drawn that the proposed approach can effectively enhance quality and edge details of the laser bubble detecting images,so that higher precision can be achieved.