Adaptive Blob Size Thresholding in Video Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video analytics systems face challenges in accurately determining the minimum blob size for effective blob filtering, leading to noisy blob detection and reduced accuracy in object detection and tracking due to the inability to distinguish between real objects and background noise.
Innovation Solution
A method and system for determining a minimum blob size threshold by analyzing the histogram of blob sizes, applying lower boundary conditions, and temporally filtering the initial minimum blob size to adaptively set the threshold for each video frame, thereby filtering out noisy blobs and enhancing detection and tracking accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed blob size threshold is used for filtering, then the system complexity is reduced, but the detection accuracy deteriorates due to inability to adapt to varying scene conditions
Solution Approach 1:
The patent implements dynamic blob size thresholding by analyzing the histogram distribution of blob sizes in each video frame and determining the lower boundary adaptively. Instead of using a fixed threshold, the system calculates the threshold based on the actual distribution characteristics of blobs in the current frame, allowing it to adapt to varying scene conditions while maintaining reasonable system complexity through automated histogram analysis.
Solution Approach 2:
The patent changes the blob size threshold parameter dynamically based on histogram analysis. By examining the distribution of blob sizes and identifying the lower boundary of the histogram, the system adjusts the threshold parameter to match the actual scene content, thereby improving detection accuracy without requiring manual intervention or complex configuration.
2Measurement precision
If a lower blob size threshold is used, then more small objects are detected, but false positives increase due to noisy blobs from background elements
Solution Approach 1:
The patent employs feedback through histogram analysis to determine the optimal blob size threshold. By analyzing the distribution of blob sizes in each frame and identifying the lower boundary of the histogram, the system receives feedback about the actual scene content and adjusts the threshold accordingly. This feedback mechanism allows the system to distinguish between real small objects and noisy blobs from background elements like trees and shadows, reducing false positives while maintaining detection accuracy.
3Object-generated harmful factors
If a higher blob size threshold is used, then false positives are reduced, but detection of real small objects is lost
Solution Approach 1:
The patent makes the blob size threshold dynamic by base it on the lower boundary of the blob size histogram for each frame. This dynamic approach allows the threshold to automatically adjust to the actual distribution of objects in the scene, ensuring that real small objects are not missed while still filtering out noisy blobs. The system adapts to different scenes by analyzing the histogram distribution rather than using a static threshold.
4Adaptability or versatility
If manual blob size threshold configuration is used, then adaptability to different scenes is improved, but ease of operation deteriorates due to complex tuning requirements
Solution Approach 1:
The patent implements self-service by automatically determining the blob size threshold through histogram analysis of each video frame. The system performs the configuration task itself by analyzing the distribution of blob sizes and identifying the lower boundary, eliminating the need for manual tuning or complex configuration. This automated approach maintains high scene adaptability while significantly improving ease of operation.
Data Source
AI summary
Techniques and systems are provided for processing video data. For example, techniques and systems are provided for determining blob size thresholds. Blob sizes of blobs generated for a video frame can be determined. A lower boundary of a category of blob sizes can then be determined that corresponds to a minimum blob size of the video frame. The lower boundary is determined from a plurality of possible blob sizes including the blob sizes of the blobs and one or more other possible blob sizes. One of the possible blob sizes is determined as the lower boundary when one or more lower boundary conditions are met by characteristics of the possible blob size. A blob size threshold for the video frame is assigned as the minimum blob size corresponding to the lower boundary.