ISAR Radar Target Recognition via Hough Transform Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radar detection methods have poor resolution in range and cross-range, making it impossible to determine the number and type of targets, limiting their utility in decision-making processes.
Innovation Solution
A multiple-target radar recognition method and apparatus that uses ISAR images processed through a series of modules including formation analysis, target separation, optimization, feature extraction, classification, and signature analysis, employing two-dimensional transforms like the Hough transform to enhance target identification and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional radar detection methods are used, then the system is simple to operate, but the resolution in range and cross-range is poor (roughly 75 m and 1000-3000 m respectively)
Solution Approach 1:
The radar trace is segmented into multiple traces by detecting local maxima in the radial velocity spectrum. This segmentation allows the system to process and analyze multiple potential targets within a single radar trace, thereby improving measurement precision without requiring a complete system overhaul
Solution Approach 2:
The invention introduces a new dimension of analysis by computing the radial velocity spectrum and identifying local maxima. This additional dimensional approach (velocity domain analysis) enables higher resolution target detection and classification without fundamentally changing the physical radar hardware
2Loss of information
If conventional radar detection methods are used, then the device complexity is low, but it is impossible to determine the number and type of targets
Solution Approach 1:
The system performs preliminary classification by computing dominant diffuser maps and vibration maps before final target identification. This preliminary analysis of structural characteristics allows the system to determine target type and number in advance, reducing information loss while managing processing complexity through staged analysis
Solution Approach 2:
The invention introduces intermediary representations (dominant diffuser maps and vibration maps) that bridge the gap between raw radar data and target classification. These intermediary structures preserve target information while enabling systematic analysis of target number and type without requiring direct complex processing of raw data
3Measurement precision
If ISAR images are processed through multiple modules, then target recognition accuracy is improved, but computational complexity increases
Solution Approach 1:
The system extracts only the most discriminative features (dominant diffusers and vibration characteristics) from the complete ISAR image data. By taking out and focusing on these key features rather than processing all image data, the system maintains high recognition accuracy while reducing computational energy requirements
Solution Approach 2:
The invention applies local quality analysis by computing dominant diffuser maps that focus on specific high-reflectivity regions of the target rather than analyzing the entire ISAR image uniformly. This localized approach concentrates computational energy on the most informative areas, improving accuracy while reducing overall computational load
Data Source
Figure 1
Figure 2~4
Figure 5a~6
AI summary
A multiple-target radar recognition method including the steps of acquiring ISAR images (I) of a monitored region; detecting moving targets (T) in the acquired ISAR images (I); and classifying the targets (T) detected in the acquired ISAR images (I). The number of targets (T) is determined by applying a two-dimensional transform to the ISAR images (I). A signature analysis of the targets (T) is performed on the basis of maps of dominant diffusers of the targets (T). As an alternative maps of vibration of the targets (T) are used.