Image Capturing Apparatus Object Detection Reliability Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image capturing apparatuses face challenges in balancing object detection accuracy and speed, with face detection offering high accuracy but slow processing, and moving-subject detection providing faster processing but lower accuracy, leading to trade-offs in auto focus and exposure control.
Innovation Solution
An image capturing apparatus with a combination of face detection and moving-subject tracking, where a reliability evaluation unit assesses the accuracy of face detection results to determine when to engage moving-subject detection for tracking, ensuring accurate and stable object tracking while preventing incorrect region tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face detection is used to detect objects from individual images by extracting face features, then detection accuracy is improved, but processing time increases and detection speed decreases
Solution Approach 1:
The patent dynamically switches between face detection mode and moving-subject detection mode based on the detected object's movement characteristics. When an object exhibits significant movement between frames, the system transitions from accurate but slow face detection to faster moving-subject detection, thereby maintaining detection speed while preserving accuracy for stationary or slowly moving objects.
Solution Approach 2:
The patent changes the detection parameter (detection method) based on the movement magnitude of the detected object. By monitoring movement between consecutive frames and adjusting the detection approach accordingly, the system optimizes the balance between detection accuracy and processing speed adaptively.
2Speed
If moving-subject detection is used to detect objects by differential operation between frame images, then processing speed is improved, but detection accuracy decreases when object luminance changes
Solution Approach 1:
The patent dynamically selects the appropriate detection method based on real-time analysis of object characteristics. When luminance changes are detected or when face features are identifiable, the system switches to face detection mode to maintain accuracy, while utilizing moving-subject detection for stable, non-luminance-changing objects to maximize speed.
Solution Approach 2:
The patent introduces an intermediary evaluation mechanism that assesses detection reliability and movement characteristics to determine which detection method to employ. This intermediary layer mediates between the two detection approaches, selecting the most appropriate method based on current conditions to optimize both speed and accuracy.
3Measurement precision
If face detection is continuously performed in EVF images to track objects, then detection accuracy is maintained, but computation complexity increases and processing time increases
Solution Approach 1:
The patent segments the detection process into two distinct modes: face detection for initial identification and verification, and moving-subject detection for continuous tracking. By dividing the continuous detection task into these segments and switching between them based on movement thresholds, the system reduces overall computational complexity while maintaining accuracy.
Solution Approach 2:
The patent applies partial face detection action by using it only when necessary (for initial detection or when movement is minimal) rather than continuously. This partial application of the more accurate but computationally intensive method reduces overall computation complexity while maintaining sufficient detection accuracy.
Data Source
AI summary
An image capturing apparatus is provided that is capable of performing both object detection using image recognition and object detection using movement detection on successively captured images. In the image capturing apparatus, the reliability of the result of the object detection using image recognition is evaluated based on the previous detection results. If it is determined that the reliability is high, execution of the object detection using movement detection is determined. If it is determined that the reliability is low, non-execution of the object detection using movement detection is determined. With this configuration, the object region can be tracked appropriately.