Receipt Image Decoding Using AI Bounding Box Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for collecting purchase data from consumers are manual and resource-intensive, burdening panelists and marketing research entities, and are prone to errors and high costs.
Innovation Solution
The use of artificial intelligence techniques such as deep learning, computer vision, and natural language processing to automatically extract and decode purchase data from images of receipts, reducing the need for manual data entry and improving data accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual methods are used to collect purchase data from consumers, then panelists can provide purchase information, but the process becomes resource-intensive, time-consuming, and error-prone
Solution Approach 1:
The patent replaces manual mechanical data entry processes with automated optical character recognition (OCR) and image processing systems. Consumers capture images of purchase documents using mobile devices, and the system automatically extracts, decodes, and processes the purchase data through machine learning models, eliminating the need for manual typing and reducing errors.
Solution Approach 2:
The system enables consumers to independently capture and submit purchase data through mobile device images without requiring researcher intervention. The automated processing pipeline handles extraction, decoding, and validation of purchase information, allowing the system to serve itself rather than relying on manual labor for data collection.
2Productivity
If manual data entry is used, then purchase data can be collected, but the cost of rewards programs increases to maintain panelist participation
Solution Approach 1:
The patent replaces resource-intensive manual data entry operations with automated image processing and machine learning systems. The OCR and decoding pipelines process purchase data automatically, dramatically reducing the time and human resources required per data point while maintaining or improving data quality.
Solution Approach 2:
The system transforms the data collection parameter from manual text input to automated image processing. By changing the input modality from typed text to captured images and implementing automated extraction algorithms, the system achieves higher productivity with reduced resource consumption.
3Productivity
If automated image processing is used to extract purchase data, then data collection efficiency improves, but the system complexity increases
Solution Approach 1:
The patent divides the complex data extraction process into distinct modular components: image capture, optical character recognition, text decoding, data validation, and processing. Each component handles a specific task independently, making the overall complex system manageable through functional segmentation and reducing the cognitive load on individual modules.
Data Source
AI summary
Methods, apparatus, systems, and articles of manufacture are disclosed that decode purchase data using an image. An example apparatus includes processor circuitry to execute machine readable instructions to at least crop an image of a receipt based on detected regions of interest, apply a first mask to a first cropped image to generate first bounding boxes corresponding to rows of the receipt, apply a second mask to a second cropped image to generate second bounding boxes corresponding to columns of the receipt, generate a structure of the receipt by mapping words detected by an optical character recognition engine to corresponding first bounding boxes and second bounding boxes based on a mapping criterion, classify the second bounding boxes by identifying an expression of interest in ones of the second bounding boxes, and generate purchase information by extracting text of interest from the structured receipt based on the classifications.