Real-Time Text Detection on Mobile Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile communications devices lack effective text input interfaces, with existing methods such as stylus-based touch screens being cumbersome and pen scanners being unsuitable for mobile use due to size and availability issues.
Innovation Solution
A mobile communications device with an integrated camera analyzes a video stream in real-time to detect text, allowing users to select and preprocess text for remote optical character recognition (OCR) via an online service, enabling efficient text input without the need for on-device OCR resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a stylus-based touch screen is used for text input, then text input capability is improved, but device cost increases and ease of operation deteriorates due to small size and two-hand requirement
Solution Approach 1:
The patent extracts the text recognition function from the input device itself and relocates it to a remote server. The mobile device only captures images via camera and transmits them, while the complex OCR processing is performed externally. This eliminates the need for expensive touch screens and stylus while maintaining text input capability.
Solution Approach 2:
The patent introduces an online service with OCR capability as an intermediary between the mobile device and the text input function. Instead of the mobile device directly performing text recognition, it uses the remote service as a mediator to convert captured images into text, resolving the contradiction between simplicity and text input capability.
2Ease of operation
If pen scanners are used for text capture, then text detection capability is improved, but portability deteriorates due to cable connection requirement and misplacement risk
Solution Approach 1:
The patent merges the text detection function directly into the mobile communication device by integrating a camera. This combination eliminates the need for separate pen scanner devices and their cable connections, making the system portable and eliminating misplacement risks while maintaining text detection capability.
Solution Approach 2:
The patent replaces the mechanical cable-connected pen scanner system with an optical camera-based system integrated into the mobile device. This substitution eliminates physical connections and improves portability while maintaining or enhancing text detection capability through digital image capture.
3Productivity
If OCR function is integrated on the mobile device, then text processing speed is improved, but device resources are overwhelmed
Solution Approach 1:
The patent segments the OCR system into two parts: a lightweight client application on the mobile device that handles image capture and transmission, and a full-featured OCR engine on the remote server. This segmentation allows fast text processing without overwhelming mobile device resources, as the complex processing is performed externally.
Solution Approach 2:
The patent moves the OCR processing from the spatial dimension (on-device hardware) to the network dimension (remote server). By transitioning from local processing to cloud-based processing, the system achieves fast text recognition without consuming mobile device resources, utilizing the network as a new dimension for resource allocation.
Data Source
AI summary
A mobile communications device with an integrated camera is directed towards text. A video stream is analyzed in real time to detect one or more words in a specified region of the video frames and to indicate the detected words on a display. Users can select a word in a video stream and subsequently move or extend the initial selection. It is thus possible to select multiple words. A subregion of the video frame comprising the detected word(s) is pre-processed and compressed before being sent to a remote optical character recognition (OCR) function which may be integrated in an online service such as an online search service.