Adaptive Thresholding for Text Recognition in Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for providing textual information to portable computing devices are inefficient, as they require manual input of phone numbers, email addresses, and web addresses, and struggle with variations in lighting and contrast, leading to inaccurate text detection.
Innovation Solution
Implementing adaptive thresholding techniques in image processing algorithms to recognize and isolate text in images, allowing portable devices to automatically detect and interpret text patterns like phone numbers, email addresses, and URLs, even under varying lighting conditions, and perform associated actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional text detection methods are used, then the system can detect text in images, but the accuracy decreases under variations in lighting and contrast
Solution Approach 1:
The patent implements adaptive thresholding that dynamically adjusts the threshold value based on local image characteristics rather than using a fixed threshold. The threshold is recalculated for different regions of the image to accommodate variations in lighting and contrast, allowing the text detection system to maintain high accuracy across diverse imaging conditions.
Solution Approach 2:
The patent applies different thresholding parameters to different local regions of the image based on their specific characteristics. By analyzing local statistics such as mean and standard deviation in each region, the system tailors the thresholding approach to match local lighting and contrast conditions, thereby improving text detection accuracy in non-uniform illumination scenarios.
2Loss of time
If manual input methods are used for providing textual information, then the system can ensure accurate data entry, but the time consumption increases
Solution Approach 1:
The patent enables the system to automatically capture and process text from images without requiring manual user input. The optical character recognition system self-services by detecting, extracting, and interpreting text from captured images, then automatically populating form fields or initiating actions based on the recognized text, thereby eliminating time-consuming manual typing while maintaining operational simplicity.
3Measurement precision
If simple thresholding is used for text detection, then the processing speed is fast, but the accuracy decreases under varying lighting conditions
Solution Approach 1:
The patent changes the parameters of the thresholding algorithm from fixed values to dynamically calculated values based on local image statistics. By computing parameters such as local mean intensity and standard deviation for each region, the system adapts the thresholding parameters to match local lighting conditions, improving accuracy without requiring overly complex processing frameworks.
Data Source
AI summary
Various approaches for providing textual information to an application, system, or service are disclosed. In particular, various embodiments enable a user to capture an image with a camera of a portable computing device. The computing device is capable of taking the image and processing it to recognize, identify, and/or isolate the text in order to forward the text to an application or function. The application or function can then utilize the text to perform an action in substantially real-time. The text may include an email, phone number, URL, an address, and the like and the application or function may be dialing the phone number, navigating to the URL, opening an address book to save contact information, displaying a map to show the address, and so on. Adaptive thresholding can be used to account for variations across an image, in order to improve the accuracy and efficiency of text recognition processes.