Chunlin Huang

3D Ground Penetrating Radar to Detect Tree Roots and Estimate Root Biomass in the Field

The objectives of this study were to detect coarse tree root and to estimate root biomass in the field by using an advanced 3D Ground Penetrating Radar (3D GPR) system. This study obtained full-resolution 3D imaging results of tree root system using 500 MHz and 800 MHz bow-tie antennas, respectively. The measurement site included two larch trees, and one of them was excavated after GPR measurements. In this paper, a searching algorithm, based on the continuity of pixel intensity along the root in 3D space, is proposed, and two coarse roots whose diameters are more than 5 cm were detected and delineated correctly. Based on the detection results and the measured root biomass, a linear regression model is proposed to estimate the total root biomass in different depth ranges, and the total error was less than 10%. Additionally, based on the detected root samples, a new index named “magnitude width” is proposed to estimate the root diameter that has good correlation with root diameter compared with other common GPR indexes. This index also provides direct measurement of the root diameter with 13%–16% error, providing reasonable and practical root diameter estimation especially in the field.

Design of bow-tie antenna with high radiating efficiency for impulse GPR

High radiating efficiency of antenna is very crucial for GPR applications. However, the bow-tie antenna, which is widely used in impulse GPR, has very low radiating efficiency because remarkable energy fed into antenna is radiated as the form of end reflection. In this paper, we study how to design bow-tie antenna with high radiating efficiency for impulse GPR. We find that, if the bow-tie antenna is excited by a bipolar pulse, the radiation efficiency can be significantly improved by utilizing the energy in end reflections. And the improvement is implemented by optimizing the antenna length to superpose the main pulse with the end reflection of a radiated pulse.

Tree roots detection based on circular survey using GPR

A circular survey using GPR is employed for tree roots detection in field measurement. A new 3D imaging algorithm is also proposed for implementing the 3D reconstruction of the subsurface targets in cylindrical coordinate system. The proposed algorithm can avoid the spatial interpolation of the acquired data between coordinate systems and employs the wave equation extrapolation technique to implement the 3D imaging efficiently. A practical GPR device for the circular survey using 400 MHz bow-tie antenna was developed for the tree root measurement. Finally, the simulated and experimental results confirmed the efficiency of the proposed circular survey and the imaging algorithm for the tree roots detection.

A kind of MIMO ground penetrating radar plane antenna array and corresponding imaging method

Antenna array is widely used in real time imaging of ground penetrating radar (GPR). With circumstance restricting, the number and interval of antenna elements are difficult to fit the need of high imaging resolution. In this paper, a kind of 4-transmitter and N-receiver MIMO GPR plane antenna array is proposed. Based on phase center approximation (PCA) theory, its equivalent antenna array is a uniform square antenna array with 4N T/R antenna elements, and the interval of equivalent elements is half of that of the real receiving elements. The equivalent uniform square antenna array has two synthetic apertures and can obtain the 2D information of target. So, the 2D image of the target can be derived by using narrowband transmitting signals and the 3D image can be derived by using wideband transmitting signals. Corresponding imaging method and procedure are also given in this paper.

A method of removing interference fringes on spherical subsurface imaging with continuous wave penetrating radar

This paper focuses on the affection of the curved surface media. In order to study it, the continuous wave penetrating radar (CWPR) is applied to detect targets buried in the spherical media. The antennae move in a plane above the spherical surface and execute two dimensions sampling. Different from planar media, this special geometry makes the echoes of the surface interfere, and several circular interference fringes which awfully degrade the imaging quality could be found. In this paper, a novel method is proposed to remove the interference fringes for improving the spherical subsurface imaging. Due to the geometric symmetry, a distribution difference in spatial spectrum domain between the echo of spherical surface and back-scattering field of target beneath the surface can be found out subsequently when a 2D Fourier transformation has been conducted on the data matrix. Therefore, a filter based on this difference is designed to remove the disturbance of interference fringes. Taking advantage of spatial spectrum filtering, CWPR successfully wipes off the fringes caused by spherical surfaces and makes substantial progress in image quality of the subsurface imaging. Moreover, the numerical simulation and experiment results validate the availability of the proposed method.

Miniature multimode deep ground penetrating radar

Deep ground penetrating radar, as its name says, is able to image the deep buried objects with high resolution. It becomes an important tool for human to know the underground world. In this paper a stepped frequency deep penetrating radar system has been developed. This miniature system is designed for three operation modes: Mode 1 is fast measurement, real-time data storage, off-line data analysis and imaging. Mode 2 is wireless real-time detection and imaging. Mode 3 is on-line real-time detection and imaging. Three modes can be switched flexibly for different working platform, such as low altitude aerocraft, planetary rover or lander, UAV (unmanned aerial vehicle), remote control vehicle and robot, vehicles or just hand holding. The miniature radar host is designed based on the embedded frame to satisfy the multi-modes operation condition including light-weight, small volume and low power consumption.

Holographic subsurface imaging for medical detection

Holographic subsurface imaging radar (HSIR) has high plane imaging resolution, and has the potential in the application of human body inspection. This paper studies mainly on the feasibility of using HSIR in the medical diagnosis. A series of experiments have been carried out to image the different parts of human body, including palm, forearm, leg, knee, chest, back, and belly. The imaging results can show the shapes of bones in body tissues clearly, but the fat and muscle are not obvious. This indicates the permittivity difference between bone and other tissues is large, and the reflectivity from bone is also high. Moreover, the imaging results easily suffer the influence from the uneven human body surface.

Improving holographic radar imaging resolution via deconvolution

Holographic radar usually adopts antenna with wide beam pattern to form a big synthetic aperture and to obtain high spatial resolution. However, previous works have indicated that the antenna pattern is essentially a low-pass filter in wavenumber domain and it reduces the practical resolution. To eliminate the influence of the antenna pattern and improve the practical resolution, this paper presents a deconvolution method implemented in two-dimension wavenumber domain. The proposed method acquires parameters by scaling and shifting the target echo to design a Wiener filter. With this Wiener filter, the received signal is deconvolved to restore the wavenumber spectrum. Some numerical simulations and laboratory experiments were conducted. The imaging results of the deconvolved data show that the resolution improved by 50%.

Inversion of Ground Penetrating Radar Data Based on Neural Networks

We present a novel inversion approach using a neural network to locate subsurface targets and evaluate their backscattering properties from ground penetrating radar (GPR) data. The presented inversion strategy constructs an adaptive linear element (ADALINE) neural network, whose configuration is related to the unknown properties of the targets. The GPR data is reconstructed (compression) to fit the structure of the neural network. The constructed neural network works with a supervised training mode, where a series of primary functions derived from the GPR signal model are used as the input, and the reconstructed GPR data is the expected/target output. In this way, inverting the GPR data is the equivalent of training the network. The back-propagation (BP) algorithm is employed for the training of the neural network. The numerical experiments show that the proposed approach can return an exact estimation for the target’s location. Under sparse conditions, an inverted backscattering intensity with a relative error lower than 3% was achieved, whereas for the multi-dominating point scenario, a higher error rate was observed. Finally, the limitations and further developments for the inverting GPR data with the neural network are discussed.

Detecting target under non-flat surface with holographic radar

The ability to penetrate medium and provide high-resolution image makes holographic radar popular among researchers in recent years. However, the strong clutter caused by the non-flat surface, which is unavoidable in practice, may obscure the image of target. This paper presents a Principal Component Analysis (PCA) based method to remove the surface clutter and enhance the imaging result. The symmetric non-flat surface and the consequent clutter are analyzed at first to reveal the low intrinsic dimensionality of the surface clutter signal in the recorded data matrix. Then the PCA is applied to project the recorded data into different subspaces and separate target response form clutter. With clutter suppressed, the target response is focused with wavenumber domain Filtered Backprojection (FBP) to obtain clear imaging result. Numerical simulations and laboratory experiments are conducted and the results validate the proposed method.