Crack Detection Using Improved Spectral Clustering Considering Effective Crack Features
Crack Detection Using Improved Spectral Clustering Considering Effective Crack Features
复制标题
DOI：
10.1145/3303714.3303750
复制
发表时间：
2018-12
期刊：
影响因子：
--
通讯作者：
Daiki Shiotsuka;Kousuke Matsushima;Osamu Takahashi
中科院分区：
文献类型：
--
作者：
Daiki Shiotsuka;Kousuke Matsushima;Osamu Takahashi
路面裂缝造成各种交通问题。因此，我们应该适当地修复它们。目前，人们提出了多种计算机视觉中的裂纹检测方法，如光谱聚类。该方法的主要优点是对道路特定噪声具有鲁棒性。然而，就处理时间而言，这种方法效率低下。本文提出了利用稀疏矩阵改进处理性能的方法。
Pavement cracks cause various traffic problems. Therefore, we should repair them properly. Nowadays a variety of crack detection methods in computer vision have been proposed, such as spectral clustering. Principal advantages of this method are robust against road-specific noises. However, the approach is inefficient in terms of processing time. In this paper, we propose improvement of processing performance by sparse matrix.