Radar Target Identification via Hough Line Feature Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing automatic target recognition systems for radar images face challenges in accurately identifying targets due to noise spikes, clutter returns, range and Doppler smearing, and atmospheric conditions, as they often rely on simulated data rather than real-world radar images.

Innovation Solution

A target identification system that projects 2-dimensional radar image features onto 1-dimensional vectors, using adaptive parameters and Hough Line processing to determine points of interest and classify targets based on length and section-specific features, allowing for dynamic identification classes tailored to the target environment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If simulated data is used to train target identification technology, then training can be performed, but the technology does not account for real-world radar image issues such as noise spikes, clutter returns, range and Doppler smearing, and atmospheric confrontations

Engineering Contradiction:
Improvetarget identification accuracyVSAvoidadaptability to real-world conditions
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system changes the parameters of training data from simulated to real-world radar images, incorporating actual noise characteristics, clutter patterns, and atmospheric effects present in operational environments. This allows the identification algorithm to adapt to real-world variations while maintaining identification accuracy.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If all known target types are attempted to be identified, then comprehensive coverage is achieved, but it is challenging, if not impossible, to correctly identify all known targets

Engineering Contradiction:
Improvetarget type coverageVSAvoididentification accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system segments the target identification process into distinct phases: feature extraction from radar images, classification into broad categories, and detailed identification within categories. This segmentation allows the system to handle diverse target types systematically while maintaining high accuracy for each category through specialized classification rules.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If feature extraction is performed on all radar image data, then complete target characterization is achieved, but the process becomes computationally intensive

Engineering Contradiction:
Improvetarget characterization accuracyVSAvoidprocessing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system extracts only the most discriminative features from radar images, such as silhouette contours, key geometric parameters, and distinctive reflector patterns. By selecting only the most informative features rather than processing all image data, the system achieves accurate target characterization while reducing computational burden and improving processing throughput.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2523016B1Target Identification for a Radar Image
Publication Date: 2016.02.03 RAYTHEON CO
  • EP2523016B1 patent drawingFigure 1
  • EP2523016B1 patent drawingFigure 2A
  • EP2523016B1 patent drawingFigure 2B

AI summary

The target identification technology described herein includes a method, a system, and a computer program product. In some examples, the system includes a length estimation module configured to determine a length of a target from a radar image based on a range profile, the radar image, and one or more adaptive parameters. The system can include a points of interest module configured to identify at least one point of interest of the target from the radar image based on the length of the target, the range profile, the radar image, the one or more adaptive parameters, and Hough Line processing. The system can include an identification module configured to determine a target identification from a plurality of identification classes based on the length of the target and the at least one point of interest of the target.