Simultaneous Character and Barcode Recognition via Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for recognizing characters and barcodes in mobile communication terminals require separate scanning and use different resolution images, leading to inefficiencies and failures in simultaneous recognition, especially when both are present in a single high-resolution image.
Innovation Solution
A method and apparatus that divide an image into text and barcode areas, allowing for simultaneous character and barcode recognition using appropriate recognition techniques and programs stored in the device's memory, thereby reducing recognition time and improving success rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate scanning methods are used for character recognition and barcode recognition, then each recognition task can be performed with dedicated processing, but the total recognition time increases and simultaneous recognition fails
Solution Approach 1:
The patent divides the image into multiple regions of interest (ROIs) based on detected barcode positions. Character recognition is performed only in specific ROIs surrounding the barcode, rather than processing the entire image. This segmentation approach reduces the computational area and time while maintaining recognition accuracy for both barcodes and characters.
2Measurement precision
If high-resolution image is used for character recognition, then character recognition accuracy is improved, but barcode recognition time increases excessively or fails
Solution Approach 1:
The patent applies different processing qualities to different regions of the image. High-resolution processing is applied only to character-containing ROIs, while barcode regions are processed at lower resolution or using dedicated barcode algorithms. This local differentiation maintains character recognition accuracy while reducing overall processing time for barcode recognition.
Data Source
AI summary
A method for controlling a character and barcode recognition apparatus by receiving an image including at least one text image and at least one barcode, dividing the received image into a text area and a barcode area, the text area including the at least one text image and the barcode area including the at least one barcode, recognizing at least one character in the text area by performing character recognition on the text area, and recognizing the at least one barcode in the barcode area by performing barcode recognition on the barcode area.