【摘要】 针对图像边缘分割方法进行桥梁裂缝检测精度不高的问题,提出基于三维激光扫描图像块分割及特征融合的龙脑桥裂缝检测技术。首先通过三维激光扫描生成图像像素序列,并采用中值滤波处理三维激光扫描图像,然后采用子空间分解方法对三维激光扫描图像进行块分割,结合邻域能量特征融合法进行裂缝统计分析,实现裂缝准确检测和标定,最后将其应用龙脑桥裂缝检测。结果表明,该方法的龙脑桥裂缝检测准确性好,定位精度高。更多还原

【Abstract】 With laser scanning laser image acquisition Borneol Bridge,combined with image processing technique for detecting cracks. Edge segmentation method for crack detection accuracy is not high question,put forward a kind of based on 3 d laser scanning image segmentation and feature fusion of borneol bridge crack detection technology. First of all to 3 d laser scanning of borneol bridge,generate the image pixel sequence,three-dimensional laser scanning image median filtering processing. Then improve subspace decomposition method is used to filter the three-dimensional laser scanning image segmentation,neighborhood energy feature fusion method is adopted to improve the selectivity of the statistical analysis of bridge cracks,the cracks in the accurate detection and calibration. Finally through the simulation performance testing,the results showed that the proposed method is adopted to improve the detection accuracy of boreal bridge cracks is good,high positioning accuracy.更多还原