Superpixel Segmentation for Extracting Multiple Documents from One Image
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods fail to effectively handle multiple documents copied, photographed, or scanned onto a single image frame, leading to manual processing and resource consumption.
Innovation Solution
A neural network-based system that generates a superpixel segmentation map, identifies connected components, and creates image dividing lines to separate documents within a single image, utilizing a downscale, context, and final classification blocks with rectifier activation functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional methods are used for document detection, then the system is simple to implement, but it cannot effectively handle multiple documents on a single image frame
Solution Approach 1:
The patent applies segmentation by dividing the image processing task into distinct stages: superpixel segmentation to group adjacent pixels, connected component analysis to identify document regions, and boundary detection to separate multiple documents. This multi-stage segmentation approach enables the system to handle multiple documents on a single image frame by breaking down the complex detection task into manageable segments.
2Productivity
If manual processing is used for multiple documents, then the system remains simple, but productivity decreases and resource consumption increases
Solution Approach 1:
The patent implements self-service by enabling the system to automatically detect, segment, and process multiple documents without human intervention. The neural network-based architecture performs autonomous document detection and separation, eliminating the need for manual processing while improving productivity and reducing time loss.
3Extent of automation
If a neural network-based system is implemented to separate multiple documents, then processing automation improves, but device complexity increases
Solution Approach 1:
The patent uses superpixel segmentation as an intermediary step between raw image input and final document separation. This intermediary process groups adjacent pixels into superpixels, simplifying the subsequent connected component analysis and boundary detection tasks. The intermediary approach enables effective automation while managing system complexity by breaking down the processing pipeline into intermediate stages.
Data Source
AI summary
System and method for document image detection, comprising: generating a superpixel binary mask associated with an input image, wherein each superpixel of the superpixel binary mask is derived from a probability characteristic reflecting a probability of the superpixel belonging to a certain object found in an input image; identifying a connected component in the superpixel binary mask; responsive to determining that a first number of pixels in a first line of the superpixel binary mask exceeds, by at least a predetermined threshold, a second number of pixels in a second line of the superpixel binary mask which is adjacent to the first line of the superpixel binary mask, utilizing the second line as a candidate image dividing line; and defining boundaries of one or more regions of interest based on a set of image dividing lines comprising the image dividing lines.