Digital Camera Object Detection Tagging for Efficient Image Filing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital camera technologies do not effectively utilize object detection information post-shooting for slide-show and filing processes, leading to inefficiencies in image handling and processing due to the large volume of image data and the need for time-consuming searches for face detection in high-pixel images.
Innovation Solution
A digital camera system that includes an object detecting unit for analyzing images and extracting subject information, such as position data and alignment data, which is then written as tag data to a storage medium, allowing for efficient use of this information in slide-show and filing processes by sorting and prioritizing detected objects based on significance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face detection is performed on high-pixel shot images for slide-show and filing processes, then accurate object identification is achieved, but processing time increases significantly
Solution Approach 1:
The patent performs face detection and extracts subject information during the shooting phase (preliminarily), storing this information in tag data. This preliminary action allows the detected face positions and identification results to be reused later during slide-show and filing processes, eliminating the need to re-detect faces in high-pixel images and thus significantly reducing processing time while maintaining accurate object identification.
2Reliability
If the entire shot image is processed for face detection in slide-show mode, then no face is missed, but processing speed decreases
Solution Approach 1:
The system performs complete face detection on the entire shot image during the shooting phase and stores the results in tag data. During slide-show mode, it directly retrieves and uses this pre-processed face information without re-processing the entire high-pixel image, thereby maintaining face detection completeness while significantly improving processing speed.
Solution Approach 2:
The patent creates a copy of the face detection results (subject information including position data) and stores it separately in tag data. This copy can be quickly accessed and used during slide-show mode without needing to re-analyze the original high-pixel image, thus maintaining reliability while enhancing productivity.
3Quantity of substance
If object detection information is not stored in tag data, then storage space is conserved, but post-shooting facilities cannot utilize detection results
Solution Approach 1:
The patent extracts only the essential subject information (such as face position data and identification results) from the complete shot image and stores only this extracted information in tag data, rather than storing the entire image or all possible analysis results. This selective extraction approach conserves storage space while still providing the necessary information for post-shooting facilities like slide-show and automatic filing functions.
Data Source
AI summary
A digital camera includes an object detecting unit which analyzes an analysis target image including a through image data taken in from an image pickup device to compose a shot of an object, a shot image, or a reduced image, based on a feature appearing in an image representing a subject of shooting, and detects at least one main object captured in the analysis target image, a subject information extracting unit which extracts, for each image of object detected by the object detecting unit, subject information including a position in a screen corresponding to the analysis target image and its expansion, and a tag data writing unit which writes the subject information, as a part of tag data corresponding to the shot image, to a storage medium, thereby enabling use of the information acquired by object detection at the time of shooting, also after the acquisition of the image.