The ultrasonic computed tomography (USCT) method is derived from the basic principles of X-ray section scanning. This method records the arriving times of ultrasonic wave between the probes and the sources to ealcul...The ultrasonic computed tomography (USCT) method is derived from the basic principles of X-ray section scanning. This method records the arriving times of ultrasonic wave between the probes and the sources to ealculate the elastic wave velocity values in the section using the arrival times. Through analyzed the distribution Of elastic wave velocity in aim area, the information of the strength and the homogeneity of the investigated zone could be got indirectly. The authors introduced the operational principle of USCT and a practical case of using this method to detect the interior defects in large scale concrete structural member. Compared with other exploration methods, this method is more efficient and accurate.展开更多
混凝土内部损伤破坏形态具有明显的离散性和随机性,内部损伤特征检测是混凝土细观研究的重要内容.针对已有混凝土结构内部损伤特征检测模型精度低的问题,提出一种特征共享双头Cascade R-CNN模型对混凝土CT图像的损伤特征进行检测.首先,...混凝土内部损伤破坏形态具有明显的离散性和随机性,内部损伤特征检测是混凝土细观研究的重要内容.针对已有混凝土结构内部损伤特征检测模型精度低的问题,提出一种特征共享双头Cascade R-CNN模型对混凝土CT图像的损伤特征进行检测.首先,为了有效识别损伤特征的空间信息,构建具有空间敏感性的fc-head(fully connected head)与空间相关性的conv-head(convolution head)相结合的Cascade R-CNN网络模型;其次,通过特征共享的方法将检测网络各层级分类信息进行融合,提升低IOU(intersection over union)阈值(0.5~0.7) ROI (regions of interest)检测任务的精度.实验结果表明,所提方法在检测混凝土CT图像的损伤特征中平均精度达到91.31%,比原始的Cascade R-CNN提高3.04%,低IOU阈值(0.5~0.7) ROI平均精度提高1.49%,该模型可以较好地从混凝土CT图像中检测出细观损伤部分,具有精度高、运算简单、易于工程实现等特点.展开更多
基金Supported by Project of the National High Technology Research and Development Program of China(No.2007AA06Z215)
文摘The ultrasonic computed tomography (USCT) method is derived from the basic principles of X-ray section scanning. This method records the arriving times of ultrasonic wave between the probes and the sources to ealculate the elastic wave velocity values in the section using the arrival times. Through analyzed the distribution Of elastic wave velocity in aim area, the information of the strength and the homogeneity of the investigated zone could be got indirectly. The authors introduced the operational principle of USCT and a practical case of using this method to detect the interior defects in large scale concrete structural member. Compared with other exploration methods, this method is more efficient and accurate.
文摘混凝土内部损伤破坏形态具有明显的离散性和随机性,内部损伤特征检测是混凝土细观研究的重要内容.针对已有混凝土结构内部损伤特征检测模型精度低的问题,提出一种特征共享双头Cascade R-CNN模型对混凝土CT图像的损伤特征进行检测.首先,为了有效识别损伤特征的空间信息,构建具有空间敏感性的fc-head(fully connected head)与空间相关性的conv-head(convolution head)相结合的Cascade R-CNN网络模型;其次,通过特征共享的方法将检测网络各层级分类信息进行融合,提升低IOU(intersection over union)阈值(0.5~0.7) ROI (regions of interest)检测任务的精度.实验结果表明,所提方法在检测混凝土CT图像的损伤特征中平均精度达到91.31%,比原始的Cascade R-CNN提高3.04%,低IOU阈值(0.5~0.7) ROI平均精度提高1.49%,该模型可以较好地从混凝土CT图像中检测出细观损伤部分,具有精度高、运算简单、易于工程实现等特点.