Weighted Background Blending for Object Movement Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional security monitoring systems often experience false alarms or missed alarms due to improper processing of previous images and inaccurate identification of moving objects, leading to a need for improved methods and devices for detecting object movement in images.
Innovation Solution
A device and method that include a weight determination circuit, an image blending circuit, and an object movement detection circuit, which determine weights for pixel positions, blend input and background images, and perform a sum of absolute difference (SAD) calculation block by block to generate moving object indication data and signals, using thresholds to enhance detection reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional image comparison methods are used to detect moving objects, then the security monitoring system can identify moving objects by comparing current images with previous images, but misjudgments (false alarms or missed alarms) occur due to improper processing of previous images and inaccurate identification
Solution Approach 1:
The patent divides the image processing into distinct segments: background image generation is separated from moving object detection. The background image is generated through weighted accumulation of historical frames, while the detection phase uses SAD calculation on the difference between current and background images. This segmentation allows each process to be optimized independently, improving both reliability and precision.
Solution Approach 2:
The patent performs preliminary action by generating and updating the background image before the actual moving object detection occurs. The background image is continuously refined using weighted accumulation of historical frames, so when detection is needed, the comparison is already optimized. This preliminary preparation reduces misjudgments during the actual detection phase.
2Productivity
If simple image comparison is used, then the processing is fast and simple, but false alarms occur due to environmental changes like light switching
Solution Approach 1:
The patent applies dynamics by using a time-varying weighting factor in the background image generation. The weight dynamically adjusts based on the age of historical frames, with more recent frames having higher weights. This dynamic weighting allows the system to adapt to environmental changes like light switching while maintaining processing efficiency, as the weighted accumulation is computationally efficient yet produces more reliable background models.
3Measurement precision
If detailed pixel-level comparison is performed, then object identification accuracy improves, but computational complexity and processing time increase
Solution Approach 1:
The patent segments the comparison process into block-level SAD calculation rather than full pixel-level comparison. By dividing the image into blocks and calculating the sum of absolute differences at the block level, the system maintains good detection accuracy while significantly reducing computational complexity. This segmented approach balances precision and complexity effectively.
Data Source
AI summary
A device for detecting the movement of object in images includes a weight determination circuit, an image blending circuit, and an object movement detection circuit. The weight determination circuit determines multiple weights according to an input image and a background image, each weight corresponding to a pixel position. The image blending circuit blends the input image and the background image based on the weights to generate an updated background image. The object movement detection circuit performs a sum of absolute difference (SAD) calculation, block by block, with the input image and the background image or the updated background image to generate a moving object indication data. The object movement detection circuit generates an object movement signal according to the moving object indication data and at least one threshold. Each block contains multiple pixels.