Image Processing Algorithm Selection for Face Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional face detection systems in digital cameras face challenges with high computational complexity and degraded performance due to unreliable input frames caused by poor image quality, leading to low detection rates and high false positive rates.
Innovation Solution
An image processing method and apparatus that selects appropriate algorithms based on image quality indicators from auxiliary sensors or processing circuits, skipping or filtering out unreliable frames to reduce computational complexity and improve detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional face detection system processes each input frame, then detection coverage is improved, but computational complexity increases
Solution Approach 1:
The patent applies the skipping principle by selectively skipping unreliable frames based on quality assessment. The system evaluates each frame's reliability using quality metrics and algorithm selection information, then skips processing for frames deemed unreliable. This allows the system to maintain detection coverage for reliable frames while reducing computational complexity by avoiding processing of poor-quality frames that would yield inaccurate results anyway.
2Productivity
If conventional face detection system processes all frames including poor quality ones, then frame processing completeness is improved, but detection accuracy deteriorates
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the processing decision based on frame quality parameters. The system uses quality metrics and algorithm selection information as parameters to determine whether to process each frame. By changing the processing parameter (process or skip) based on the quality parameter assessment, the system maintains high detection accuracy for reliable frames while avoiding accuracy degradation from processing poor-quality frames.
3Adaptability or versatility
If algorithm selection is performed for each source image, then processing adaptability is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-generating algorithm selection information and quality metrics before the actual face detection processing. The system prepares the algorithm selection information in advance based on the source image characteristics, so that when detection is needed, the appropriate algorithm is already selected. This preliminary preparation enables processing adaptability without adding significant complexity during the critical detection phase, as the selection work is done beforehand.
Data Source
AI summary
An exemplary image processing method includes the following steps: receiving an image input composed of at least one source image; receiving algorithm selection information corresponding to each source image; checking corresponding algorithm selection information of each source image to determine a selected image processing algorithm from a plurality of different image processing algorithms; and performing an object oriented image processing operation upon the source image based on the selected image processing algorithm. The algorithm selection information indicates an image quality of each source image and is generated from one of an auxiliary sensor, an image processing module of an image capture apparatus, a processing circuit being one of a video decoder, a frame rate converter, and an audio/video synchronization (AV-Sync) module, or is a user-defined mode setting.