Photograph Classification Using Image Similarity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing photo album management systems are inefficient in organizing and storing large numbers of similar photographs, leading to redundant images and inconvenience in viewing and management.
Innovation Solution
A method and apparatus for photograph classification and storage that involves obtaining photograph time information, extracting image characteristics, matching and calculating similarity between photographs, and storing or marking similar images in the same set, with adjustments based on photograph time intervals and image quality determination using deep neural network training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If photographs are arranged in a given sequence based on size or photograph time, then the organization is simple and systematic, but a large number of redundant photographs are shown on the same page, causing inconvenience to users when viewing and managing the photographs
Solution Approach 1:
The patent segments photographs into different groups based on similarity analysis. Photographs with high similarity are placed in the same group, while those with low similarity are separated into different groups. This segmentation reduces the number of redundant photographs displayed on the same page while maintaining systematic organization through group-based structure.
Solution Approach 2:
The patent changes the organization parameter from simple chronological or size-based ordering to similarity-based grouping. By introducing similarity as a new parameter for organization, the system can identify and group redundant photographs together, allowing users to view fewer representative images while maintaining access to all photographs through the grouping structure.
2Reliability
If multiple similar photographs are stored and displayed separately, then all photographs are preserved, but the server workload increases and user viewing efficiency decreases
Solution Approach 1:
The patent merges similar photographs into the same group based on similarity analysis. Photographs that are determined to be redundant or highly similar are combined into single groups, with one representative photograph displayed while others are stored in the same group. This merging reduces server workload for displaying and processing photographs while preserving all original images in their respective groups.
3Loss of information
If all photographs are displayed on the same page, then complete information is provided to users, but the complexity of photograph management and viewing increases significantly
Solution Approach 1:
The patent introduces a new dimension of organization by grouping photographs based on similarity. Instead of displaying all photographs in a single flat sequence, the system creates a hierarchical structure with groups as the first dimension and individual photographs within groups as the second dimension. This dimensional change allows users to navigate through grouped photographs more efficiently while maintaining access to complete information.
Data Source
AI summary
A method and an apparatus for photograph classification and storage by matching an image characteristic of a first photograph with an image characteristic of a second photograph in a directory, and calculating the similarity between the first photograph and the second photograph and presenting the first photograph and the second photograph in a front-end page as located in a same subdirectory when the similarity between the first photograph and the second photograph is larger than a preset threshold. The beneficial effects being that a number of similar images in a user's photo album can be sorted efficiently and placed into the same directory to facilitate the user's management and viewing of the photographs.