Multi-frame Moving Object Detection in Low SNR Imagery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional remote sensing image analysis systems are not robust in low signal-to-noise ratio (SNR) environments, making it difficult to distinguish real objects from noise in images.
Innovation Solution
A computing system that identifies low SNR target pixels and candidate motion paths in multiple frames of remote sensing image data, performs filtering and normalization to mitigate background elements and noise, and generates a sum image chip to determine actual object motion by accumulating signal energy along predicted paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional image analysis systems are used in low SNR environments, then object detection can be performed, but the system cannot reliably distinguish real objects from noise
Solution Approach 1:
The patent transitions from analyzing single image frames to analyzing sequences of multiple image frames along candidate motion paths. By adding the temporal dimension and examining pixel intensity variations across multiple frames at predicted object locations, the system can distinguish real objects (which maintain consistent intensity patterns along their motion paths) from random noise (which exhibits erratic, inconsistent patterns). This dimensional expansion from 2D spatial analysis to 3D spatio-temporal analysis resolves the contradiction between detection reliability and signal discrimination precision in low SNR environments.
2Measurement precision
If multiple image frames are analyzed along candidate motion paths, then object detection accuracy improves, but computational complexity increases
Solution Approach 1:
The patent performs preliminary actions by first identifying candidate motion paths using a motion model before conducting detailed intensity analysis. The system generates predicted object locations along these candidate paths in advance, then uses these predictions to guide the extraction and comparison of image chips from multiple frames. This preliminary motion estimation reduces the search space and focuses computational resources only on relevant regions and time frames, thereby improving detection accuracy while controlling computational complexity.
3Measurement precision
If image chips are extracted and summed along motion paths, then sub-pixel precision detection is achieved, but processing time increases
Solution Approach 1:
The patent replaces traditional mechanical or brute-force image processing methods with a model-based approach. Instead of exhaustively searching all possible object positions and orientations, the system uses a motion model to predict object locations and extracts image chips only at these predicted positions along candidate motion paths. This model-guided extraction and summation process achieves sub-pixel precision through intensity pattern analysis while significantly reducing processing time compared to comprehensive search methods.
Data Source
AI summary
A plurality of remote sensing images of a scene are received. A potential target object can be identified in one of the images, wherein the target object has a low signal-to-noise ratio (SNR). A candidate motion path of the target object can be generated based upon the images. A predicted position of the target object along the candidate motion path is determined for each of the remote sensing images. An image chip is extracted from each of the images, where each image chip is centered about the predicted position of the target object in its corresponding image. A sum image chip is generated based upon the image chips. An indication that the potential target object is an actual object in the images is output based upon a value of a center pixel of the sum image chip.