In the paper, first the Algorithm of optimum threshold is used to remove the noise pixels of the pavement image, such as pavement marking image; Next the pavement image is divided into 400 subimages which has 100 x 100 pixels, the feature of pavement subimages are represented by gray variance, and the threshold of the good pavement subimages feature is found though the method of online learning, then the sub-images are made to do the image binarization. Finally, the counting feature of binary pavement image is used as the parameter to classify the pavement distress images, and searching pavement surface distress images are achieved.
The Automatic Search of Pavement Surface Distress Image Based on On-Line Learning
First International Conference on Transportation Engineering ; 2007 ; Southwest Jiaotong University, Chengdu, China
2007-07-09
Conference paper
Electronic Resource
English
The Automatic Search of Pavement Surface Distress Image Based on On-Line Learning
British Library Conference Proceedings | 2007
|Automatic inspection of pavement cracking distress
British Library Online Contents | 2006
|Wiley | 1975
|Wavelet-Based Pavement Distress Classification
Transportation Research Record | 2005
|