Machine-Readable Object Detection Using Stable Scene Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Consumer devices with limited processing capabilities face computational inefficiencies when detecting and tracking machine-readable objects in images, leading to increased battery usage and diverted processing power.
Innovation Solution
A method that analyzes image frames for sharpness and stability before submitting them to a machine-readable object detector, using a stable scene detector to filter out frames unlikely to contain the object, and a region tracker to efficiently track the object's location using translation metrics from motion sensor data or image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If every image frame is analyzed by the machine-readable object detector, then detection accuracy is improved, but processing power consumption increases
Solution Approach 1:
The patent applies preliminary action by performing pre-analysis of image frames using a stable scene detector and sharpness detector before submitting frames to the machine-readable object detector. This preliminary filtering identifies frames that are likely to contain machine-readable objects and have sufficient image quality, thereby avoiding unnecessary processing of unsuitable frames and reducing overall power consumption while maintaining detection accuracy for relevant frames
Solution Approach 2:
The patent extracts and processes only the essential characteristics of image frames (sharpness, stability, motion metrics) before full detection. By separating the filtering stage from the detection stage, the system extracts only those frames meeting specific criteria for further analysis, reducing the workload on the computationally expensive machine-readable object detector
2Measurement precision
If every image frame is analyzed by the machine-readable object detector, then detection completeness is improved, but device performance degradation increases
Solution Approach 1:
The system performs preliminary filtering using motion metrics, sharpness analysis, and stability detection before full object detection. This preliminary action prepares frames by identifying those with high probability of containing machine-readable objects, ensuring detection completeness for relevant frames while preventing device performance degradation through selective processing
Solution Approach 2:
The detection process is segmented into multiple stages: motion metric calculation, sharpness detection, stability detection, and finally machine-readable object detection. Each stage processes frames differently based on their characteristics, with only frames passing all preliminary stages being submitted to the computationally intensive detection algorithm, thus maintaining completeness while preserving device performance
3Measurement precision
If image frames with motion blur or instability are processed, then detection coverage is improved, but false negative rate increases
Solution Approach 1:
The patent performs preliminary action by detecting image sharpness and frame stability before submission to the object detector. Frames exhibiting motion blur or instability are identified and excluded from processing, ensuring that only high-quality frames are analyzed. This maintains detection coverage for suitable frames while significantly reducing the false negative rate caused by processing degraded images
4Measurement precision
If all image frames are submitted for detection, then detection thoroughness is improved, but battery life decreases
Solution Approach 1:
The system performs preliminary action through motion metric calculation, sharpness detection, and stability detection to identify frames worthy of full processing. By filtering out frames unlikely to contain machine-readable objects before submission to the detector, the system maintains thorough detection of relevant frames while dramatically reducing battery consumption associated with processing all frames
Solution Approach 2:
The patent applies partial action by processing only a subset of frames that meet specific criteria rather than all frames. The preliminary detectors identify and select only those frames with appropriate motion metrics, sharpness, and stability for machine-readable object detection, performing partial processing that maintains thoroughness for relevant cases while conserving battery life
Data Source
AI summary
A method to improve the efficiency of the detection and tracking of machine-readable objects is disclosed. The properties of image frames may be pre-evaluated to determine whether a machine-readable object, even if present in the image frames, would be likely to be detected. After it is determined that one or more image frames have properties that may enable the detection of a machine-readable object, image data may be evaluated to detect the machine-readable object. When a machine-readable object is detected, the location of the machine-readable object in a subsequent frame may be determined based on a translation metric between the image frame in which the object was identified and the subsequent frame rather than a detection of the object in the subsequent frame. The translation metric may be identified based on an evaluation of image data and/or motion sensor data associated with the image frames.