The Detection Of Airport Runways And Aircrafts Based On PolSAR Image

Polarimetric Synthetic Aperture Radar(PolSAR) can achieve day/night and all-weather surveillance in the region of interest(ROI). Meanwhile, it can access to the polarization scattering matrix, which reflects the amplitude and phase information of targets through transmitting and receiving different polarized signals. Therefore, comparing with the single-channel and single-polarization SAR., it presents a better capability of gaining surface features and has been widely used in many fields such as agriculture, forestry, geology,oceanography, and military detection.Airport is an important military and civilian facilities, and the automatic detection of it has important practical value in many fields, such as military reconnaissance, precision strike,emergency rescue, aircraft navigation and so on. As a typical man-made objects,aircraft is one of the main targets of military reconnaissance. So the detection of aircraft research has the vital significance. In this paper, the detection of airport runways and aircrafts under complex scenes of Pol SAR image is studied respectively.For the detection of airport runways under complex scenes, two new algorithms are proposed. In the first algorithm, firstly, training samples templates are selected according to H/? classification with the prior of backward scattering mechanism of the terrain. Then, the statistical characteristic of polarized coherent matrix are used to classify the image and get suspected areas. Finally, the airport runway region is identified with the size and structural characteristics of runways. The second algorithm is an improved algorithm. It uses h/q classification to extract templates, and adds span detector(SD) to discriminate runways.Multi-look fully PolSAR datasets acquired by U.S.UAVSAR systems are used to verify the efficiency of the new algorithms. The experimental results show that the two novel algorithms can both detect runways effectively. Beside,the latter method is more efficient than the former one. Not only because a more intact structure of the runways, clearer outline and a lower false alarm rate is owned, but also the complexity of calculation is reduced.There are two aspects research about the aircrafts detection based on PolSAR. One the one hand, for large and complex scenes, according to the feature that the aircrafts are usuallyparked in the airport tarmac, taxiway and other regions, a new aircrafts detection approach based on a priori knowledge is proposed. This method using the Shannon entropy to discriminate aircrafts and runways after detecting the region of airport runways. On the other hand, for small and complex scenes, a new aircarfts detection method based on the conditional entropy and Shannon entropy is put forward. The method is under the premise that airport area has been positioned, and it using contional entropy, Shannon entropy and the structure feature of the aircrafts to achieve multi-target detection. Multi-look fully PolSAR datasets acquired by U.S.UAVSAR systems and AIRSAR systems are used to verify the efficiency of the two new methods. The experimental results show that the two novel methods can both detect aircrafts effectively.