Hierarchical partition clustering method based on DBSCAN to solve the depth value of Contaminated runway

To solve the error caused by using laser measurement technology to continuously sample the movement of pollutants covering the road surface and classify the high points to measure the depth of the pollutants on the road surface, a density-based spatial clustering based on segmented clustering is proposed ( DBSCAN) algorithm for detection and calibration of pavement pollutant depth measurement data.First, according to the segmentation detection distance constraint and category parameters, the ε neighborhood radius and the MinPts minimum number of neighborhood points are selected to cluster and group the pollutant depth data set. Then the similar depth clusters within the segmentation interval are merged. Then the DBSCAN1D algorithm is used to Calculate the correlation of depth data in the clustered data set, and finally select the midpoint as the segmented depth measurement value.Through an example verification on the depth movement sampling data of pollutants on an airport pavement, the results show that the method proposed in this paper can effectively avoid the impact of collecting fluctuating noise data and measurement error data while ensuring that the segmented detection distance constraints are met, and ensure the runway surface The accuracy of the pollutant depth measurement data and the detection effect are better than the K-means method. This method can provide airports with more accurate runway surface condition information, providing strong support for ensuring flight safety.