AR Text Clarity via Virtual Surface Overlay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Head-mounted displays (HMDs) face challenges in providing clear pass-through views, particularly with text clarity, which affects user safety and usage time due to the unclear details in real-space images captured by their cameras.
Innovation Solution
An image displaying method that captures real-space images, detects text regions, recognizes text content, obtains feature points, and creates a virtual surface for displaying text content, enhancing text clarity through feature extraction and virtual image generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the camera directly captures real space images for pass through views, then the device complexity is low and operation is simple, but the text clarity and detail visibility deteriorate
Solution Approach 1:
The patent segments the real space image into multiple regions including text regions and non-text regions. By detecting text regions specifically and processing them separately through OCR and virtual surface generation, the system enhances text clarity without unnecessarily processing the entire image, thus managing complexity effectively.
Solution Approach 2:
The patent introduces intermediate processing steps including text region detection, OCR recognition, and virtual surface creation as mediators between the original captured image and the final displayed pass through view. These intermediaries transform unclear text into clear virtual text overlays while preserving the original background image.
2Loss of information
If the camera directly captures real space images for pass through views, then the processing time is short and response is fast, but the text readability and user safety deteriorate
Solution Approach 1:
The patent extracts only the text regions from the complete real space image for enhanced processing. By identifying and isolating text areas through text region detection, the system applies OCR and virtual surface generation only to these extracted text portions, minimizing processing time while maximizing text information clarity.
Solution Approach 2:
The patent performs preliminary text region detection and OCR recognition on captured images before displaying the pass through view. By pre-processing and identifying text regions in advance, the system prepares clear text overlays that can be quickly composited onto the final display, reducing overall processing time while ensuring text readability for user safety.
Data Source
AI summary
An image displaying method is disclosed. The image displaying method includes the following operations: capturing a first image of a real space by a camera based on a first viewing direction when the camera is located at a first camera position, wherein the first image includes a first text image; detecting a text region according to the first image by a processor, wherein the text region includes the first text image; recognizing the first text image to obtain a first text content by the processor; obtaining several first feature points of the text region according to the first image by the processor; creating a first virtual surface according to the several first feature points by the processor; and displaying a first virtual image with the first text content appending to the first virtual surface by a display circuit.