首页>
外国专利>
- FREQUENCY-WAVENUMBER ANALYSIS METHOD AND APPARATUS THROUGH DEEP LEARNING-BASED SUPER RESOLUTION GROUND PENETRATING RADAR IMAGE GENERATION
- FREQUENCY-WAVENUMBER ANALYSIS METHOD AND APPARATUS THROUGH DEEP LEARNING-BASED SUPER RESOLUTION GROUND PENETRATING RADAR IMAGE GENERATION
展开▼
机译:- 频率 - 波数分析方法和设备通过深层学习的超分辨率地面穿透雷达图像生成
展开▼
页面导航
摘要
著录项
相似文献
摘要
A deep learning-based ultra-high-resolution surface-penetrating radar image generating frequency-wavenumber analysis apparatus according to an embodiment of the present invention includes: a memory in which a deep learning-based surface-penetrating radar data processing program is stored; and a processor executing the program stored in the memory, wherein the processor excites the electromagnetic pulse wave for each channel on the surface according to the execution of the program, and measures a reflected signal corresponding to the excited electromagnetic pulse wave. For the low-resolution GPR data obtained from the scanner, ultra-high-resolution GPR data is generated using preset training data through the ultra-high-resolution data generation network, and the generated ultra-high-resolution GPR data is subjected to time and Transforms from the spatial domain to the frequency and wavenumber domains, performs filtering on the ultra-high-resolution GPR data in the transformed frequency and wavenumber domains, and converts the filtered ultra-high-resolution GPR data into time and Reconstruct the ultra-high-resolution GPR data by inverse transform to the spatial domain.
展开▼