Text-Image Correspondence via Morphological Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to correctly match article body text with images when multiple images are provided for a single article, leading to confusion in understanding the content, especially in limited screen displays like smartphones.
Innovation Solution
An information processing device performs morphological analysis to divide text and captions into morphemes, generates caption abstracts, and calculates correlations to determine the correct correspondence between phrases of the article body text and images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all images are displayed similarly, then the user can see all images, but it becomes difficult to understand which image corresponds to the current sentence
Solution Approach 1:
The patent applies asymmetry by displaying the image corresponding to the currently displayed or read-out sentence in a different style (emphasized display) compared to other images. This creates a visual distinction that helps users understand which image corresponds to the current text, resolving the confusion that arises when all images are displayed uniformly.
2Area of stationary object
If only one image is displayed on a small screen, then the display area is utilized efficiently, but it becomes difficult to display multiple images that may correspond to different parts of the article
Solution Approach 1:
The patent applies dynamics by making the displayed image change dynamically based on which sentence is currently being displayed or read-out. The system automatically switches between different images as the user progresses through the article, allowing multiple images to be displayed sequentially over time while maintaining efficient use of the limited screen space.
3Device complexity
If correspondence is determined at the document level, then the process is simple, but the correspondence between specific sentences and images is inaccurate
Solution Approach 1:
The patent applies segmentation by dividing the article into individual sentences and determining correspondence between each sentence and relevant images. This sentence-level segmentation approach provides much more precise matching compared to document-level correspondence, ensuring that the displayed image accurately corresponds to the specific sentence being read or displayed.
4Measurement precision
If correspondence is determined for each sentence, then the matching precision is high, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-processing the article and images to extract key features and establish potential correspondences before the actual reading or display process. This preliminary preparation reduces the computational burden during real-time operation, allowing for precise sentence-level matching without excessive processing delays.
Data Source
AI summary
An information processing device (10) includes: a morphological analysis unit (11a, 11b) performing morphological analysis to divide each of an article body text included in an article and a caption of each of images into morphemes; a phrase acquiring unit (12) dividing the article body text into phrases on a basis of a result of the morphological analysis performed by the morphological analysis unit (11b); and a correspondence determining unit (13). The correspondence determining unit (13) determines correspondence between each of the phrases of the article body text and the images by calculating a correlation between the caption and each of the phrases of the article body text on a basis of the result of the morphological analysis performed by the morphological analysis unit (11a).