二阶矩&灰度差分的桥梁裂缝快速识别方法
|
摘要
传统基于摄像的桥梁裂缝识别,存在实时性与准确性的矛盾。结合桥梁病害图的颜色和纹理特征,利用图像质量评价方法,提出一种基于二阶矩与灰度差分的桥梁裂缝安全评价方法。选择桥梁质量较好时的图像组成参考图像,间隔采样建立评价图像序列。提出基于颜色矩与灰度差分的桥梁裂缝安全评价算法。通过颜色与纹理在HSV颜色空间的安全评价函数,计算评价图像与参考图像之间的相关程度,获得评价参数序列。评价参数大于阈值时,说明图像上存在显著变化,提示人工检测。实验在6组桥梁裂缝图上进行,数据表明,评价结果与算子提取的病害程度一致,同时在裂缝识别上优于峰值信噪比等方法。
The traditional recognition of bridge cracks based on camera has the contradiction between real-time and accuracy. Combining the color and texture features of bridge disease maps, this paper proposed a safety assessment method of bridge crack based on second-order moment and gray difference by using image quality assessment method. The images with better quality of bridge were artificial selected as reference image, and the sampling interval was used to form evaluation image sequence. We proposed a bridge crack safety evaluation algorithm based on color moment and gray difference. Through the safety evaluation function of color and texture in HSV color space, the correlation degree between the evaluation image and reference image was calculated, and the evaluation parameter sequence was obtained. When the evaluation parameters were larger than the threshold value, there are significant changes in the image, suggesting manual detection. The experiment was carried out on six groups of bridge crack maps. The data show that the evaluation results are consistent with the disease degree extracted by operator, and are superior to peak signal to noise ratio in crack identification.
The traditional recognition of bridge cracks based on camera has the contradiction between real-time and accuracy. Combining the color and texture features of bridge disease maps, this paper proposed a safety assessment method of bridge crack based on second-order moment and gray difference by using image quality assessment method. The images with better quality of bridge were artificial selected as reference image, and the sampling interval was used to form evaluation image sequence. We proposed a bridge crack safety evaluation algorithm based on color moment and gray difference. Through the safety evaluation function of color and texture in HSV color space, the correlation degree between the evaluation image and reference image was calculated, and the evaluation parameter sequence was obtained. When the evaluation parameters were larger than the threshold value, there are significant changes in the image, suggesting manual detection. The experiment was carried out on six groups of bridge crack maps. The data show that the evaluation results are consistent with the disease degree extracted by operator, and are superior to peak signal to noise ratio in crack identification.
引文
[1] Yang X,Sun Q,Wang T.Image quality assessment improvement via local gray-scale fluctuation measurement[J].Multimedia Tools and Applications,2018,77(18):24185-24202.
[2] Pei S C,Chen L H.Image Quality Assessment Using Human Visual DOG Model Fused With Random Forest[J].IEEE Transactions on Image Processing,2015,24(11):3282-3292.
[3] Lin C,Chen Z,Liao N.Full-reference quality assessment for stereoscopic images based on binocular vision model[C]//2016 Visual Communications and Image Processing(VCIP).IEEE,2017:1-4.
[4] 管非凡,郁梅,宋洋,等.高动态范围图像客观质量评价方法[J].计算机应用,2017,37(3):695-698.
[5] 李娜,周蓬勃,耿国华,等.色彩复原图像的质量评价方法[J].计算机应用,2016,36(6):1673-1676.
[6] Sheikh H R,Bovik A C.Image information and visual quality[J].IEEE Transactions on image processing,2006,15(2):430-444.
[7] 褚江,陈强,杨曦晨.全参考图像质量评价综述[J].计算机应用研究,2014,31(1):13-22.
[8] 胡义坦,曹杰.基于边缘特征和亮度的彩色图像质量评价[J].电子设计工程,2014,22(1):156-158.
[9] Moorthy A K,Bovik A C.Blind Image Quality Assessment:From Natural Scene Statistics to Perceptual Quality[J].IEEE Transcation Image Processing,2011,20(12):3350-3364.
[10] 刘雪超,吴志勇,黄德天,等.结合自然图像统计和空域变换的无参图像质量评价[J].计算机辅助设计与图形学学报,2015(2):249-255.
[11] 吕风杰,信科.基于小波域峰态值的无参考噪声图像评价算法[J].计算机工程与应用,2017,53(7):202-205.
[12] 赵旦峰,王博,李明.基于空域特征的无参考图像质量评价算法[J].计算机工程与应用,2016,52(12):216-220.
[13] 张薇,蔡浩,胡波,等.基于解析稀疏表示的图像模糊无参考快速评价算法[J].计算机应用研究,2016,33(10):3169-3172.
[14] 徐琳,陈强,汪青.色彩熵在图像质量评价中的应用[J].中国图象图形学报,2015,20(12):1583-1592.
[15] 王冠军,吴志勇,云海姣,等.结合图像二次模糊范围和奇异值分解的无参考模糊图像质量评价[J].计算机辅助设计与图形学学报,2016,28(4):654-659.
[2] Pei S C,Chen L H.Image Quality Assessment Using Human Visual DOG Model Fused With Random Forest[J].IEEE Transactions on Image Processing,2015,24(11):3282-3292.
[3] Lin C,Chen Z,Liao N.Full-reference quality assessment for stereoscopic images based on binocular vision model[C]//2016 Visual Communications and Image Processing(VCIP).IEEE,2017:1-4.
[4] 管非凡,郁梅,宋洋,等.高动态范围图像客观质量评价方法[J].计算机应用,2017,37(3):695-698.
[5] 李娜,周蓬勃,耿国华,等.色彩复原图像的质量评价方法[J].计算机应用,2016,36(6):1673-1676.
[6] Sheikh H R,Bovik A C.Image information and visual quality[J].IEEE Transactions on image processing,2006,15(2):430-444.
[7] 褚江,陈强,杨曦晨.全参考图像质量评价综述[J].计算机应用研究,2014,31(1):13-22.
[8] 胡义坦,曹杰.基于边缘特征和亮度的彩色图像质量评价[J].电子设计工程,2014,22(1):156-158.
[9] Moorthy A K,Bovik A C.Blind Image Quality Assessment:From Natural Scene Statistics to Perceptual Quality[J].IEEE Transcation Image Processing,2011,20(12):3350-3364.
[10] 刘雪超,吴志勇,黄德天,等.结合自然图像统计和空域变换的无参图像质量评价[J].计算机辅助设计与图形学学报,2015(2):249-255.
[11] 吕风杰,信科.基于小波域峰态值的无参考噪声图像评价算法[J].计算机工程与应用,2017,53(7):202-205.
[12] 赵旦峰,王博,李明.基于空域特征的无参考图像质量评价算法[J].计算机工程与应用,2016,52(12):216-220.
[13] 张薇,蔡浩,胡波,等.基于解析稀疏表示的图像模糊无参考快速评价算法[J].计算机应用研究,2016,33(10):3169-3172.
[14] 徐琳,陈强,汪青.色彩熵在图像质量评价中的应用[J].中国图象图形学报,2015,20(12):1583-1592.
[15] 王冠军,吴志勇,云海姣,等.结合图像二次模糊范围和奇异值分解的无参考模糊图像质量评价[J].计算机辅助设计与图形学学报,2016,28(4):654-659.