Web Page Region Selection Overlay for Mixed Content Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for selecting a region on a web page displayed in a browser are limited to element-by-element selection, preventing users from choosing desired regions that combine image and text elements.
Innovation Solution
An image processing apparatus and method that allows users to select a region of a web page by inserting a region selection object with adjustable coordinates and size, enabling the extraction and printing of desired regions through a structured-document printing module, which generates intermediate data for image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If element-by-element selection is used in the browser, then the selection process is simple and compatible with structured document elements, but the user cannot select arbitrary regions such as parts of image or text elements
Solution Approach 1:
The patent introduces an intermediary component (the region selection object or overlay) that mediates between the user's selection action and the structured document elements. This overlay allows users to select arbitrary regions by drawing rectangles over the displayed web page, while the system translates these visual selections into structured data extractions. The intermediary layer bridges the gap between simple visual selection and complex structured document processing.
Solution Approach 2:
The patent segments the selection process into two independent stages: first, visual region selection through drawing rectangles on the displayed web page; second, extraction and processing of the selected region's content. This segmentation allows the selection interface to be simple and visual, while the backend processing handles the complexity of extracting content from structured documents, images, and mixed regions.
2Adaptability or versatility
If arbitrary region selection is enabled, then users can select desired regions including parts of images and texts, but the selection mechanism becomes more complex
Solution Approach 1:
The patent creates a visual copy or overlay of the web page display that serves as the selection interface. This overlay is a separate layer that mirrors the visual appearance of the web page but allows for free-form region selection. By working with this visual copy rather than directly manipulating the structured document elements, the system enables arbitrary region selection without complicating the underlying document structure or requiring complex coordinate transformations.
3Productivity
If element-by-element selection is used, then the selection process is fast and straightforward, but the user cannot combine multiple element types (image and text) in a single selection
Solution Approach 1:
The patent adds a visual dimension to the selection process by allowing users to draw rectangles over the displayed web page. Instead of selecting elements through the structured document hierarchy (one dimension), users can select regions visually based on their spatial position on the screen (another dimension). This visual coordinate system allows seamless combination of different element types (images, texts, mixed content) within a single rectangular selection, while the system efficiently translates these visual selections back to the structured document for extraction.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An image processing apparatus includes a determination unit configured to determine coordinate of a region in a web page having a plurality of elements in a structured document, the display region being specified by a user, an extraction unit configured to extract image data corresponding to the coordinate determined by the determination unit in the web page, an output unit configured to output an image based on the image data extracted by the extraction unit, wherein, when the region specified by the user is a part clipped out of an element in the web page, the extraction unit extracts image data of the part of the element corresponding to the coordinate determined by the determination unit