Imaged Target Detection Verification Using Multi-Image Repetition Criteria
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current inspection methods for detecting damage or anomalies on vehicles and other objects are often inaccurate and inefficient, as they rely on single-image inputs, which can lead to false positives due to reflections or other non-target features, and do not effectively verify the repetition of target detection regions across multiple images.
Innovation Solution
A system that processes multiple overlapping images from different angles and positions to verify target detection regions by applying detection repetition criteria, classifying candidate regions as verified or non-verified based on their appearance in multiple images, thereby improving detection accuracy and filtering out false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single-image inspection methods are used, then the inspection process is simple and fast, but detection accuracy is low due to false positives from reflections and non-target features
Solution Approach 1:
The patent transitions from single-image (2D) inspection to multi-image (temporal dimension) inspection by capturing images at different time points and comparing them. This dimensional expansion allows the system to distinguish true targets from false positives by checking for temporal consistency, thereby improving detection accuracy without requiring complex multi-sensor setups.
Solution Approach 2:
The system performs preliminary actions by capturing multiple images before making a final detection decision. By acquiring a sequence of images and pre-processing them to identify candidate target regions, the system prepares sufficient data in advance to accurately distinguish true targets from false positives, improving measurement precision while maintaining reasonable process complexity.
2Reliability
If multiple images are processed to verify target detection, then false positives are reduced, but processing time and computational resources increase
Solution Approach 1:
The patent segments the detection process into distinct stages: candidate region identification in the first image, verification by checking corresponding regions in subsequent images, and final classification. This segmentation allows the system to process only relevant regions rather than entire images, reducing computational load and processing time while maintaining high detection reliability through multi-image verification.
Solution Approach 2:
The system applies partial verification by checking only the candidate target regions identified in the first image against subsequent images, rather than performing exhaustive comparison of all image content. This partial action approach maintains detection reliability for critical targets while significantly reducing processing time and computational resource requirements.
3Measurement precision
If strict detection repetition criteria are applied, then false positives are filtered out, but true targets may be missed
Solution Approach 1:
The patent implements dynamic detection criteria that adapt based on the number of matching images and the consistency of target characteristics across the image sequence. Rather than applying fixed rigid thresholds, the system adjusts verification requirements dynamically, allowing true targets that appear consistently across multiple images to be detected while filtering out false positives that do not meet the adaptive criteria, thus balancing precision and completeness.
Data Source
AI summary
A system configured for verification of imaged target detection performs the following: (a) receive information indicative of at least two images of object(s). This information comprises candidate target detection region(s), indicative of possible detection of a target associated with the object(s). Images have at least partial overlap. The candidate region(s) appears at least partially in the overlap area. The images are associated with different relative positions of capturing imaging device(s) and of an imaged portion of the object(s). (b) process the information to determine whether the candidate region(s) meets a detection repetition criterion, the criterion indicative of repetition of candidate region(s) in locations of the images that are associated with a same location on a data representation of the object(s). (c) if the criterion is met, classify the candidate region(s) as verified target detection region(s). This facilitates output of an indication of the verified region(s).