Image Processing Device for Automatic Album Creation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
General camera users face challenges in creating wedding ceremony albums that match the quality and flow of professional albums, as existing image processing technologies fail to automatically select and arrange images similar to those in model albums while maintaining the event's chronological order.
Innovation Solution
An image processing device and method that detects similar images from user captures by comparing them to sample images, grouping them based on similarity and chronological order, and automatically arranging them to match the flow of a professional model album.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If images are automatically selected based on similarity to model album images, then the album quality approaches professional level, but images may be attached at positions that do not match the event flow
Solution Approach 1:
The patent divides the image matching process into two independent segments: (1) similarity-based matching using image recognition to find images with similar content to model album images, and (2) rank-based ordering using metadata (time, date, sequence) to arrange images in event flow order. This segmentation allows each segment to optimize for its specific purpose without interfering with the other, resolving the contradiction between image quality and flow accuracy.
Solution Approach 2:
The patent introduces rank information (derived from metadata such as time stamps, dates, and sequence numbers) as an intermediary element that mediates between the similarity-based image selection and the final album arrangement. The rank acts as a bridge that preserves event flow information while allowing flexible image selection based on similarity, enabling both professional-quality image selection and correct chronological ordering.
2Stability of the object's composition
If multiple image grouping methods are used to maintain event flow, then the album structure improves, but the complexity of image processing increases
Solution Approach 1:
The patent applies preliminary action by pre-ranking all candidate images based on their metadata (time, date, sequence) before the actual album creation process. This pre-ranking establishes the event flow order in advance, so that during album generation, images only need to be placed according to their pre-determined ranks rather than requiring complex real-time analysis of event sequences, thereby simplifying the overall processing complexity.
Solution Approach 2:
The patent changes the parameter used for image arrangement from complex scene understanding and event recognition to simple numerical rank values derived from metadata. By transforming the arrangement criterion into a quantitative parameter (rank) that can be directly compared and sorted, the system achieves stable album structure without requiring complex processing logic.
3Manufacturing precision
If professional album creation methods are applied to user albums, then album quality improves, but the time and effort required increases significantly
Solution Approach 1:
The patent implements self-service by enabling the album creation system to automatically perform image selection, similarity matching, and chronological ordering without requiring user intervention. The system uses the user's own metadata (time stamps, dates, sequence numbers) to automatically determine the event flow and arrange images accordingly, eliminating the need for users to manually curate or sequence images while still achieving professional-quality results.
Solution Approach 2:
The patent uses copying by replicating the structure and arrangement logic of professional model albums and applying it to user images. The system copies the ranking and ordering methodology from professional album creation processes and applies it automatically to user photos, allowing users to obtain professional-quality albums without investing the time and effort that would be required to create them manually.
Data Source
AI summary
Provided are an image processing device, an image processing method, and an image processing program that can decide a user image that is similar to a plurality of model images and that follows a flow of the plurality of model images. Combinations of (I2) and (P3), (I51) and (P21), (I150) and (P49), (I198) and (P79), (I201) and (P30), (I249) and (P99), (I300) and (P101), and (I301) and (P112) in which similarities between the model images and the user images are equal to or more than a first threshold value are detected. The combination of (I201) and (P30) in which a relative time difference of the imaging time is equal to or more than a second threshold value is eliminated, and the rest combinations are second combinations. From among the second combinations, a model image group and a user image group, which are a collection of similar images, are decided, and the user images similar to the model images attached on a model album are decided from the user image group. A user album is created from the decided user images.