Discriminative Color Feature Extraction for Moving Object Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image forming systems face challenges in accurately distinguishing moving objects from non-moving objects and backgrounds, especially in complex environments where the color appearance of moving objects is similar to other image areas, making it difficult to extract discriminative color features for effective object detection and tracking.
Innovation Solution
An image forming system that extracts a discriminative color feature by comparing the color vector of an object's histogram with that of the background, adjusting weights to emphasize differences, and using these weights to generate a final color model for improved object localization and tracking, incorporating a device and method that divides images into rectangles, calculates bin weights, and re-weights color histogram bins to enhance discriminative features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a representative color feature (color histogram) is used to represent the moving object, then the color distribution of the object can be captured, but the ability to discriminate the moving object from background and other objects deteriorates when color appearances are similar
Solution Approach 1:
The patent divides the object area into multiple sub-areas (e.g., head, body, limbs) and extracts color histograms for each sub-area separately. This segmentation allows the system to identify discriminative color features from specific parts of the object that differ from the background, even when the overall color distribution is similar. The segmented approach enables selective weighting of sub-area histograms based on their discriminative power.
Solution Approach 2:
The patent applies different weighting factors to different sub-area color histograms based on their local discriminative quality. Sub-areas with color distributions that are more distinct from the background are assigned higher weights, while sub-areas with similar colors to the background receive lower weights. This local quality differentiation enhances the overall discrimination capability by emphasizing informative regions.
2Loss of information
If the color histogram of the whole object is used, then the complete color distribution is represented, but the discriminative power is reduced when parts of the object have colors similar to the background
Solution Approach 1:
The patent extracts discriminative color features by separating the object into sub-areas and selectively emphasizing those sub-areas with high discriminative power. Instead of using the complete color histogram of the whole object, the system extracts and weights individual sub-area histograms, taking out only the most informative color distribution components for object discrimination.
Solution Approach 2:
The patent changes the parameter weighting in the color histogram representation by assigning different importance weights to different sub-area histograms. This parameter adjustment transforms the complete color distribution into a weighted combination that emphasizes discriminative regions while de-emphasizing non-discriminative regions, thereby improving measurement precision for object discrimination.
3Ease of operation
If standard color model schemes are used for tracking, then object detection can be performed, but tracking accuracy deteriorates in complex environments with similar color appearances
Solution Approach 1:
The patent introduces dynamic weighting factors that are calculated based on the discriminative power of each sub-area color histogram. These weights are not fixed but are dynamically adjusted according to the specific scene and object characteristics. This dynamic adaptation allows the tracking system to maintain high accuracy in complex environments by automatically emphasizing the most discriminative color features in each situation.
Solution Approach 2:
The patent employs a feedback mechanism where the discriminative power of each sub-area is evaluated and used to adjust the weighting of its color histogram. This feedback loop continuously optimizes the color feature representation based on the actual discrimination performance, thereby improving tracking accuracy in complex environments while maintaining ease of operation through automated adjustment.
Data Source
AI summary
Provided are an apparatus and method of extracting a discriminative color feature, and an image forming system including: a photographing device to photograph an image of an object; a color feature extracting device receiving the image from the photographing device, extracting a discriminative color feature from the image, generating a final color model of the object, extracting a color blob of the object based on the final color model, performing blob analysis on the extracted color blob, and generating parameters to control a posturing of the photographing device according to the blob analysis; a control device receiving from the color feature extracting device the parameters to control the posturing of the photographing device, and controlling the posturing of the photographing device; a storage device storing the photographed image of the object; and a display device displaying the photographed image of the object.