Study On The Algorithm Of Ground Penetrating Radar (GPR) To Detect Shallow Goal

Ground penetrating radar (GPR) is an effective tool to detect the underground target by using the discontinuity of underground medium. Make it have been widely used in national defense, public security, highway, railway, bridge, tunnel, airport construction, and exploration, geological survey and so on. This paper mainly studies the ground penetrating radar detection algorithms for shallow underground targets, the main contents are:First of all, this paper has a brief introduction and forth for some applications of ground penetrating radar, understand the current situation on the ground penetrating radar studies at home and abroad, and explain and elaborate the basic working principle of ground penetrating radar (GPR), the form of ground-penetrating radar echo data, model and the resolution of the ground penetrating radar (GPR), these theories belong to the basics, can lay theoretical basis for the practical application, the gain processing of the GPR echo signal is introduced, the GPR echo signal due to the attenuation factor, which leads to the signal is not very strong, so it carried out the gain processing and improve the amplitude of echo signal,and carried out clutter suppression to the two-dimensional original image, use the means method and adaptiveering to preprocess of the original 2D Radar image, processed images are verified, it is concluded that the average method in ground level off, can be very good for clutter suppression, and adaptive filtering method in the uneven ground clutter suppression can also effectively, and the length of the RLS filter is selected and verified, and it is found that the inhibition effect is best when the length of the RLS filter is 6. According to the quantitative method, a new method to measure the suppression of clutter is proposed. This method can directly calculate the specific signal to clutter ratio, and can directly see the effect of various algorithms on clutter suppression.Secondly, since the ground penetrating radar azimuth resolution is not enough high, need synthetic aperture imaging for pre-processed image, this paper improves the microwave holography SAR algorithm and propose a new algorithm, through the verification and analysis of experimental results, it is concluded that the proposed algorithm in a single target, not only can greatly improve the azimuth resolution, but also can improve processing speed, saving processing time. For the multi objectives, the proposed method is roughly equivalent to the traditional method, but the computation speed is much faster than the traditional algorithm.Finally, carry out object feature extraction and classification for the detected target, pick out the desired objectives, this paper first expound the traditional LDA, and improve the LDA algorithm, use kernel function characteristic to distinguish the target, and verified by experiments, the proposed method can be very good to distinguish the target. In this paper, the innovation points and deficiencies are analyzed, and the future research direction is discussed.