Adaptive Illuminance Filter for Video Foreground Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video surveillance systems face challenges in accurately distinguishing foreground objects due to environmental illumination changes, leading to false-positive and false-negative classifications, as they often incorrectly identify shadows and highlights as foreground or background pixels.
Innovation Solution
A method that extracts a foreground image from a video frame using a background model image to determine an approximated reflectance component, removing pixels with reflectance values below a threshold to filter out false-positive foreground pixels caused by environmental illumination effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional video surveillance systems suppress environmental illumination effects based on luminance differences, then false-positive foreground pixels are reduced, but false-positive and false-negative classifications increase due to incorrect assumption about chromaticity consistency
Solution Approach 1:
The patent segments the image formation process into distinct components: illuminance (lighting conditions) and reflectance (surface properties). By separating these components mathematically, the system can analyze foreground pixels based on reflectance characteristics rather than being misled by illumination variations, thereby improving classification accuracy without the precision losses of conventional luminance-based methods
Solution Approach 2:
The patent introduces a reflectance component as an intermediary variable that mediates between the observed image data and the actual object properties. This intermediary allows the system to distinguish true foreground objects from false positives by examining reflectance characteristics that remain consistent across different lighting conditions
2Object-affected harmful factors
If the system uses luminance-based suppression to remove shadows and highlights, then environmental illumination effects are reduced, but foreground objects may be incorrectly removed due to false-positive classifications
Solution Approach 1:
The patent changes the fundamental parameter used for analysis from luminance to reflectance. By transforming the image data to separate illuminance and reflectance components, the system can identify foreground objects based on their reflectance properties rather than brightness, eliminating the harmful effect of illumination variations while preserving true foreground objects
3Device complexity
If the system assumes false-positive pixels differ only in luminance values, then processing complexity is reduced, but measurement precision deteriorates leading to incorrect foreground detection
Solution Approach 1:
The patent adds a new dimension to the analysis by introducing reflectance as a separate component alongside luminance. This dimensional expansion allows the system to distinguish between illumination effects and object properties more accurately, improving measurement precision while maintaining manageable complexity through mathematical decomposition
Data Source
AI summary
Techniques are disclosed for removing false-positive foreground pixels resulting from environmental illumination effects. The techniques include receiving a foreground image and a background model, and determining an approximated reflectance component of the foreground image based on the foreground image itself and a background model image which is used as a proxy for an illuminance component of the foreground image. Pixels of the foreground image having approximated reflectance values less than a threshold value may be classified as false-positive foreground pixels and removed from the foreground image. Further, the threshold value used may be adjusted based on various factors to account for, e.g., different illumination conditions indoors and outdoors.