Split Image Processing Pipeline for Credit Card Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing technologies face challenges in accurately detecting and recognizing credit card information due to difficulties in focusing, perspective corrections, and computational costs, which limit user freedom and increase the complexity of the recognition process, especially when using portable devices with embedded cameras.
Innovation Solution
A split image processing pipeline operating in both high-resolution and low-resolution modes concurrently, allowing for flexible card positioning and using a Convolutional Neural Network (CNN) for character sequence recognition without explicit segmentation, to enhance detection and recognition of credit card information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the camera operates in low-resolution mode to reduce computational costs, then processing speed improves, but measurement precision deteriorates
Solution Approach 1:
The patent divides the image processing into two segments: a low-resolution path for rapid object detection and a high-resolution path for detailed character recognition. The low-resolution path quickly identifies potential credit card locations, while the high-resolution path processes only those specific regions, combining speed with precision without requiring full high-resolution processing of the entire image.
Solution Approach 2:
The system applies different quality levels to different regions of the image. Full high-resolution processing is applied only to detected credit card regions, while the rest of the image is processed at low resolution. This local quality approach ensures measurement precision is maintained where needed while preserving overall processing speed.
2Measurement precision
If the user is guided to frame the credit card precisely close to the camera to obtain sufficient image detail, then measurement precision improves, but ease of operation deteriorates
Solution Approach 1:
The system dynamically adjusts processing based on detected object characteristics. Instead of requiring static precise framing, the system can detect credit cards at various distances and orientations, then adaptively apply perspective correction and selective high-resolution processing to extract readable information, giving users freedom in positioning.
Solution Approach 2:
The system changes processing parameters based on detected conditions. When a credit card is detected at varying distances or angles, the system adjusts focal length, applies perspective transformation, and selectively processes regions at high resolution to compensate for suboptimal framing, maintaining measurement precision without constraining user positioning.
3Adaptability or versatility
If the camera focuses on objects at widely-varying distances to increase adaptability, then adaptability improves, but measurement precision deteriorates
Solution Approach 1:
The patent introduces a processing dimension to compensate for optical limitations. When the camera captures images at widely-varying distances with potentially compromised focus, the system applies computational methods including deconvolution, sharpness enhancement, and selective high-resolution processing to recover and enhance image quality, effectively adding a post-capture focus adjustment dimension.
4Measurement precision
If full high-resolution image processing is used to improve recognition accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent extracts only the necessary high-resolution processing to specific detected regions of interest (credit cards). Instead of processing the entire high-resolution image, the system identifies credit card locations in low-resolution mode, then extracts and processes only those specific regions at high resolution, reducing computational complexity while maintaining recognition accuracy.
Solution Approach 2:
The system applies high-resolution processing partially—only to detected credit card regions rather than the entire image. This partial action provides sufficient recognition accuracy for the specific task while avoiding the excessive computational complexity of processing all image data at full resolution.
Data Source
AI summary
Differing embodiments of this disclosure may employ one or all of the several techniques described herein to utilize a “split” image processing pipeline, wherein one part of the “split” image processing pipeline runs an object-of-interest recognition algorithm on scaled down (also referred to herein as “low-resolution”) frames received from a camera of a computing device, while the second part of the “split” image processing pipeline concurrently runs an object-of-interest detector in the background on full resolution (also referred to herein as “high-resolution”) image frames received from the camera. If the object-of-interest detector detects an object-of-interest that can be read, it then crops the object-of-interest out of the “high-resolution” camera buffer, optionally performs a perspective correction, and/or scaling on the object-of-interest to make it the desired size needed by the object-of-interest recognition algorithm, and then sends the scaled, high-resolution representation of the object-of-interest to the object-of-interest recognition algorithm for further processing.