| The information of road especially the damage road plays an important role in the emergency rescue when the natural disasters happened. It is of great significance in the disaster emergency rescue work. With the support of UAV's information gathering technology and the aerial remote sensing technology, the emergency rescue work has acquired new breakthrough in recent years.Aerial remote sensing provides a large number of high resolution images for realtime disaster emergency rescue, and the road to the disaster area always get much attention because it relating to people rescue. The extraction of the damage road information to the disaster area has become an important research direction in the field of emergency rescue technology. The road information is easily affected by the background and noise, it remains a difficult problem in the field of automatic target recognition. This paper studied the existing methods of the road network extraction and road damage information extracting technology. And based on the road's actual situation and the needing of the emergency rescue, two new algorithms is developed:(1) A road information extraction method is proposed. Based on the strip and direction feature of the road, road information extraction is fulfilled by matching with a vector template. Firstly, the road's vector template is established according to the geometric characteristics of the road, and then the whole image is split out too many blocks for analyzing. In the end, the road is connect into a network throughout the whole image. This method extracts road information of the disaster area, and provides information about the accurate edges and the center line of the road efficiently, which can provide support for the road damage information extracting.(2) A “line scan” methods for extracting road damage information is proposed. The information is extracted by using a vector line scan the cross section of road. Firstly, road cross-section gray matrix(one dimensional array) is marked artificially. Secondly, the line which is perpendicular to the central line if the road and whose length equals to the width of the toad, is set by the array. Thirdly, scanning the line along the central line of the road by a certain step, and marked the abnormal point, thus, the road destruction information is extracted. This method can extract the changing information about the road surface in detail, and providing information support for judging the damage degree of road of road accurately. |
PDF Full Text Request