CN-121995376-A - Terahertz imaging method based on CEEMD noise reduction

Abstract

The invention belongs to the technical field of terahertz imaging, and relates to a terahertz imaging method based on CEEMD noise reduction. The method comprises the steps of (1) decomposing sampling data in a fast time dimension by CEEMD, calculating first P IMF components, subtracting the first P IMF components from an original signal, (2) decomposing the three-dimensional data processed in the step (1) into four groups of data which are uniformly sampled in an XY plane, (3) respectively carrying out imaging processing on the four groups of data by using an RMA algorithm based on a wave number domain to form an image, and (4) carrying out coherent accumulation on four images formed by the four groups of data to form a final image. The method can reduce the calculation complexity and improve the calculation efficiency.

Inventors

• HU ZHENG
• LIU MING
• ZHOU HAORAN
• ZHANG TING
• HAN SHUNLI
• GAO TIANFENG
• CHEN FEIYU
• LIU YIXUAN
• ZHANG GUIMING
• ZHU JUNFENG
• DU ZHIQIANG
• JIANG JINCHUN

Assignees

• 中国电子科技集团公司第四十一研究所

Dates

Publication Date

20260508

Application Date

20251224

Claims (5)

1. 1. The terahertz imaging method based on CEEMD noise reduction is characterized by comprising the following steps of: (1) The sampling data on the fast time dimension is decomposed by CEEMD to calculate the first P IMF components; (2) Decomposing the three-dimensional data processed in the step (1) into four groups of data which are uniformly sampled on an XY plane; (3) Respectively carrying out imaging processing on the four groups of data by using an RMA algorithm based on a wave number domain to form an image; (4) And coherently accumulating four images formed by the four groups of data to form a final image.
2. 2. The CEEMD noise reduction-based terahertz imaging method according to claim 1, wherein four groups of equally spaced equivalent phase center sequences are found for a non-uniformly distributed sparse array to perform a wave number domain-based RMA algorithm process.
3. 3. The CEEMD noise reduction-based terahertz imaging method according to claim 1 or 2, wherein in step (3), a fast fourier transform is applied to the echo data when an RMA algorithm process is performed.
4. 4. The CEEMD noise reduction-based terahertz imaging method according to claim 1 or 2, wherein in step (4), the same position of the target in the image is identified at the time of accumulation, and the accumulation is performed with the position as an alignment center.
5. 5. The method according to claim 1 or 2, wherein the method is applicable to sparse MIMO arrays.

Description

Terahertz imaging method based on CEEMD noise reduction Technical Field The invention belongs to the technical field of terahertz imaging, and relates to a terahertz imaging method based on CEEMD noise reduction. Background Terahertz waves (Terahertz, THz) are positioned between microwaves and infrared rays in an electromagnetic spectrum, have optical and electromagnetic characteristics, have the characteristics of strong penetrability, low photon energy and the like, and have the frequency range covering 0.1-10 THz, so that terahertz imaging is widely researched in academia and industry, and has wide development and application prospects in the fields of nondestructive detection, safety inspection and the like. The combination of the terahertz wave and the MIMO-SAR (Synthetic Aperture Radar) technology becomes a key technology in the field of terahertz imaging, on one hand, the space freedom degree can be fully utilized, the equivalent phase center which is far more than the number of actual antennas can be obtained through the transmission and the reception of a plurality of antennas, and on the other hand, the high-resolution imaging capability can be obtained through the synthetic aperture processing. Existing methods use empirical mode decomposition (EMPIRICAL MODE DECOMPOSITION, EMD) to reduce noise in the signal, but suffer from modal aliasing. An integrated empirical mode decomposition (EEMD) introduces white noise into the signal to be decomposed, which improves the modal aliasing of the EMD decomposition, but the EEMD may leave some white noise in each of the intrinsic mode function (INTRINSIC MODE FUNCTION, IMF) components. Complementary empirical mode decomposition (CEEMD) applies the same white noise, but of opposite sign, to the signal, which can cancel the residual noise of the IMF component. However, CEEMD is computationally intensive in the decomposition process. Meanwhile, for sparse non-uniformly arranged arrays, if a back projection imaging algorithm is adopted, the calculation complexity is too high. Disclosure of Invention In order to solve the problems in the prior art, the invention provides a terahertz imaging method based on CEEMD noise reduction. The technical scheme provided by the invention is that the terahertz imaging method based on CEEMD noise reduction comprises the following steps: (1) Decomposing the digital signal received in the fast time dimension by CEEMD to calculate the first P IMF components; (2) Decomposing the three-dimensional data processed in the step (1) into four groups of data which are uniformly sampled on an XY plane; (3) Respectively carrying out imaging processing on the four groups of data by using an RMA method based on a wave number domain to form an image; (4) And coherently accumulating four images formed by the four groups of data to form a final image. Preferably, four groups of equivalent phase center sequences distributed at equal intervals are found for the unevenly distributed sparse array to be subjected to RMA algorithm processing based on the wave number domain. Preferably, in step (4), the same position of the target in the image is identified and accumulated with the position as the alignment center. Preferably, in step (3), when the RMA algorithm processing is performed, a fast fourier transform is used for echo data. Preferably, the method is applicable to sparse MIMO arrays. Compared with the prior art, the invention has the following beneficial effects: (1) When CEEMD is adopted for modal decomposition, only the first few IMFs are decomposed, so that the calculation complexity is reduced; (2) Four groups of equivalent phase center sequences with equal intervals are found for the unevenly distributed sparse array, so that the RMA algorithm can be utilized for rapid calculation, and the calculation efficiency is improved. Drawings FIG. 1 is a schematic diagram of a sparse MIMO array layout; Fig. 2 is a schematic diagram of an equivalent phase center distribution. Detailed Description For a clearer understanding of the technical content of the present invention, a scheme thereof will be described in detail with reference to the accompanying drawings and specific embodiments. The drawings show a preferred embodiment of the invention, but the scope of protection is not limited thereto, and can be flexibly adjusted according to specific use requirements in practical applications. The present embodiment is intended to assist those skilled in the art in comprehensively and deeply understanding the inventive concept of the present solution. The sparse MIMO array of the present invention consists of 32 transmit 12 receive antennas with coordinates (-0.055+0.01 m, -0.017,0), m=0, 2..11. The coordinates of the transmitting antennas are (Tn, 0.017,0), n=1,..32, the transmitting antennas consist of four groups of antennas, each group consisting of 8 array elements, the coordinates of the 32 transmitting antennas on the X-axis (coordinates unit: me