Real-Time Character Recognition via Preview Image Auto Focus
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional character recognition methods in portable terminals are inefficient for rapid recognition of consecutive words in real-time, as they require users to take still images and perform processing sequentially, leading to delays in providing recognition results.
Innovation Solution
A method and apparatus that activate a camera in preview mode, control auto focus for clear images, and perform character recognition using real-time preview images, excluding non-character data and utilizing a dictionary database for rapid and accurate word recognition, allowing for continuous character detection and movement detection to optimize processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional character recognition method is used (taking still image and performing processing sequentially), then character recognition can be performed, but recognition speed is slow and cannot recognize consecutive words in real time
Solution Approach 1:
The patent implements continuous character recognition by processing characters in real-time as they appear in the preview image, rather than requiring complete still images. The system continuously captures and processes character data from video frames, enabling uninterrupted recognition of consecutive words without sequential processing delays.
Solution Approach 2:
The system performs preliminary actions by pre-processing and filtering character data from preview images before final recognition. It continuously monitors and prepares character data in real-time, so when characters need to be recognized, the processing is already partially complete, reducing overall recognition time.
2Measurement precision
If character recognition is performed on all captured data, then comprehensive recognition is achieved, but non-character data increases processing complexity and time
Solution Approach 1:
The patent extracts only character data from the preview image, separating it from non-character data such as backgrounds, objects, and other visual elements. By filtering and extracting only relevant character information, the system reduces processing complexity while maintaining comprehensive character recognition accuracy.
Solution Approach 2:
The system applies different processing qualities to different parts of the image data. Character regions receive detailed recognition processing, while non-character regions are quickly filtered or ignored. This localized processing approach maintains high accuracy for characters while reducing overall computational complexity.
3Manufacturing precision
If still image capture is used for character recognition, then clear character data can be obtained, but real-time continuous recognition is not achieved
Solution Approach 1:
The patent converts discrete still image capture into continuous video frame processing. By continuously capturing and processing preview images in real-time, the system maintains image quality necessary for recognition while enabling continuous throughput of multiple characters and words without sequential bottlenecks.
Solution Approach 2:
The system transitions from static still image processing to dynamic real-time video processing. The preview image stream provides continuous, dynamic updates of character data, allowing the system to adapt to moving or changing text while maintaining recognition accuracy through real-time processing.
Data Source
AI summary
A method and an apparatus for recognizing characters using an image are provided. A camera is activated according to a character recognition request and a preview mode is set for displaying an image photographed through the camera in real time. An auto focus of the camera is controlled and an image having a predetermined level of clarity is obtained for character recognition from the images obtained in the preview mode. The image for character recognition is character-recognition-processed so as to extract recognition result data. A final recognition character row is drawn that excludes non-character data from the recognition result data. A first word is combined including at least one character of the final recognition character row and a predetermined maximum number of characters. A dictionary database that stores dictionary information on various languages using the first word is searched, so as to provide the user with the corresponding word.