Search

CN-122017829-A - Radar three-dimensional imaging method and device based on area array angular domain decomposition

CN122017829ACN 122017829 ACN122017829 ACN 122017829ACN-122017829-A

Abstract

The invention provides a radar three-dimensional imaging method and device based on area array angular domain decomposition, which are characterized by acquiring original two-dimensional area array data, carrying out transformation processing to obtain first data, carrying out angular domain decomposition on the first data, dividing an overall view field into more than two local angular domain subareas, constructing a corresponding local angular domain decomposition weight function for each local angular domain subarea, carrying out weighting processing on the first data by utilizing the local angular domain decomposition weight function to obtain second data of each local angular domain subarea, carrying out local wave number domain three-dimensional imaging processing on each second data to obtain a three-dimensional sub-image of the corresponding local angular domain subarea, and carrying out fusion on the three-dimensional sub-images of all the local angular domain subareas to obtain a final three-dimensional imaging result. And the method has the advantages of maintaining the high efficiency of wave number domain processing, simultaneously carrying out decomposition modeling on propagation differences in different angular domain areas, and improving the three-dimensional imaging quality in near field, large field of view and complex geometric scenes.

Inventors

  • DU SHAOYAN

Assignees

  • 上海辅量成像技术有限公司

Dates

Publication Date
20260512
Application Date
20260414

Claims (10)

  1. 1. The radar three-dimensional imaging method based on area array angular domain decomposition is characterized by comprising the following steps: acquiring original two-dimensional area array data, and performing transformation processing to obtain first data; Performing angular domain decomposition on the first data, dividing the whole view field into more than two local angular domain sub-regions, constructing corresponding local angular domain decomposition weight functions for each local angular domain sub-region, and performing weighting processing on the first data by using the local angular domain decomposition weight functions to obtain second data of each local angular domain sub-region; Carrying out three-dimensional imaging processing on each second data in a local wave number domain to obtain a three-dimensional sub-image of a corresponding local angular domain sub-region; and fusing the three-dimensional sub-images of all the local angular domain sub-regions to obtain a final three-dimensional imaging result.
  2. 2. The method of claim 1, wherein collecting the raw two-dimensional area array data and performing a transformation process to obtain the first data comprises: Acquiring original two-dimensional area array data, and transforming the original two-dimensional area array data to a distance frequency domain through fast Fourier transform to obtain first data; And/or the original two-dimensional area array data is two-dimensional area array data or two-dimensional virtual area array data of the vehicle-mounted radar.
  3. 3. The method of claim 1 or 2, wherein angular domain decomposing the first data, dividing the overall field of view into more than two local angular domain sub-regions comprises: Dividing the whole view field into more than two local angular domain sub-areas according to the azimuth view field range and the pitch view field range, wherein each local angular domain sub-area has a corresponding central direction; And/or constructing a corresponding local angular region decomposition weight function for each of the local angular region sub-regions comprises: for the first The local angular domain sub-regions are used for constructing a local angular domain decomposition weight function as follows: Wherein, the Is a positive integer; Windowing functions for the array; Is an imaginary unit; is the center wave number; is the first Unit vectors in the center direction of the sub-areas of the local angular domain; Is the position coordinates of the array elements, Indicating the horizontal orientation of the array, Representing the vertical position of the array.
  4. 4. The method of claim 1, wherein weighting the first data with the local angular domain decomposition weight function comprises: multiplying the local angular domain decomposition weight function with the first data; and/or the fusion is coherent fusion, incoherent fusion or self-adaptive weighted fusion, and the weight of the self-adaptive weighted fusion is set according to local signal-to-noise ratio, focusing definition, spectral energy distribution or main lobe sidelobe index.
  5. 5. The method of claim 1, wherein separately performing a local wavenumber domain three-dimensional imaging process on each second data to obtain a three-dimensional sub-image of the corresponding local angular domain sub-region comprises: Performing Fourier transform on each second data in a two-dimensional array dimension to obtain a two-dimensional space frequency spectrum in a corresponding local angular domain sub-region; According to the mapping relation of the local wave number domain, mapping the two-dimensional space spectrum to a three-dimensional wave number domain to obtain three-dimensional wave number domain data of non-uniform sampling; Carrying out regularized resampling on the three-dimensional wave number domain data of the non-uniform sampling to obtain three-dimensional wave number domain data on a regular grid; And performing three-dimensional inverse Fourier transform on the three-dimensional wave number domain data on the regular grid to obtain a three-dimensional sub-image of the corresponding local angular domain sub-region.
  6. 6. The method of claim 5, wherein the local wave number domain mapping relationship is: Wherein, the And Is a two-dimensional spatial frequency of the frequency domain, Is of frequency The number of waves corresponding to the number of waves, Is the wave number in the depth direction.
  7. 7. A radar three-dimensional imaging device based on area array angular domain decomposition for the method of any of claims 1-6, comprising: the acquisition unit is used for acquiring original two-dimensional area array data and performing conversion processing to obtain first data; The decomposition unit is used for performing angular domain decomposition on the first data, dividing the whole view field into a plurality of local angular domain sub-areas, constructing corresponding local angular domain decomposition weight functions aiming at each local angular domain sub-area, and performing weighting processing on the first data by utilizing the local angular domain decomposition weight functions to obtain second data of each local angular domain sub-area; The processing unit is used for respectively carrying out three-dimensional imaging processing on each second data in the local wave number domain to obtain a three-dimensional sub-image of the corresponding local angular domain sub-region; and the fusion unit is used for fusing the three-dimensional sub-images of all the local angular domain sub-areas to obtain a final three-dimensional imaging result.
  8. 8. An electronic device comprising a memory and a processor, the memory having stored thereon a program executable on the processor, which when executed by the processor, causes the electronic device to implement the method of any of claims 1-6.
  9. 9. A readable storage medium having a program stored therein, characterized in that the program, when executed, implements the method of any one of claims 1-6.
  10. 10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method of any of claims 1-6.

Description

Radar three-dimensional imaging method and device based on area array angular domain decomposition Technical Field The invention relates to the technical field of radar signal processing, in particular to a radar three-dimensional imaging method and device based on area array angular domain decomposition. Background The vehicle millimeter wave radar has the advantages of all-weather operation, strong rain and fog resistance, low cost, easy integration and the like, and is widely applied to the fields of intelligent driving, parking assistance, road side sensing, space environment reconstruction and the like. As the perception task is gradually developed from traditional point target detection to target contour restoration, obstacle three-dimensional positioning, road boundary modeling and space occupation estimation, the radar system has higher requirements on three-dimensional imaging capability. Especially in vehicle-mounted forward-looking and around-looking scenes, the target is always in a medium-short distance range, the spatial distribution is complex, and the requirement of high-precision environment perception is difficult to meet only by relying on the traditional distance-azimuth two-dimensional imaging. In the existing vehicle-mounted radar three-dimensional imaging technology, more common methods comprise a two-dimensional array-based azimuth-elevation beam forming method, a back projection-based three-dimensional imaging method and a wave number domain reconstruction method by reference to SAR/near field scanning imaging ideas. The wave number domain imaging method has good engineering application prospect because the wave number domain imaging method can convert a space propagation model into a frequency domain processing problem and realize high-efficiency imaging by means of FFT, interpolation and inverse transformation. However, for vehicle millimeter wave radars, especially systems employing two-dimensional area arrays or two-dimensional MIMO virtual arrays, three-dimensional imaging still faces the following objective difficulties: Firstly, the two-dimensional area array simultaneously carries space sampling information of horizontal dimension and vertical dimension, and the target echo has coupling among azimuth angle, pitch angle and distance dimension. If a unified wave number domain model is directly built on the whole two-dimensional aperture, model mismatch is easy to occur under the conditions of large view field, near field or large strabismus. Secondly, the aperture of the vehicle-mounted radar array is limited, the mounting posture is restrained, the target distance span is larger, the propagation characteristics in different angular domain areas are inconsistent, the reconstruction of the three-dimensional wave number domain with uniform full aperture is directly carried out, and the local focusing degradation, the sidelobe lifting and the spatial resolution reduction are easily caused. And if a full three-dimensional space voxel-level back projection method is adopted to avoid model mismatch, the operand is obviously increased, and the vehicle-mounted real-time processing requirement is difficult to meet. In view of the above difficulties, the following disadvantages are prevalent in the prior art: first, conventional wave number domain three-dimensional imaging methods typically default to an array response that satisfies a uniform propagation model throughout the field of view, without considering model bias in forward looking wide field of view and near field scenes. Secondly, the traditional two-dimensional area array processing method generally directly performs two-dimensional FFT or unified interpolation on all array element data in a unified way, and cannot correct or reconstruct the propagation differences of different angular areas in a partitioning way. Thirdly, under the near field and large field conditions, only a single global wave number domain mapping relation is relied on, so that obvious focusing errors and imaging distortion can occur in the field edge area. Fourth, although the imaging effect can be improved by the partial high-precision method, large-scale three-dimensional search or voxel-by-voxel coherent accumulation is generally required, so that the calculation complexity is high, and the vehicle-mounted real-time deployment is not facilitated. Therefore, a radar three-dimensional imaging method and device based on area array angular domain decomposition are needed to improve the above problems. Disclosure of Invention The invention aims to provide a radar three-dimensional imaging method and device based on area array angular domain decomposition, which can perform decomposition modeling on propagation differences of a two-dimensional area array in different angular domain areas while maintaining the advantage of high efficiency of wave number domain processing, so that the three-dimensional imaging quality in near field, large field of view and co