Document Image Synthesis for Fold Distortion Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image reading technologies fail to adequately suppress distortions in documents with folds or curls, leading to reduced accuracy in recognition processes such as handwritten character recognition and image recognition due to incomplete distortion suppression when holding down the document with hands or fingers.
Innovation Solution
An image processing apparatus that includes an image input unit, an operation detecting unit, a partial-image acquiring unit, and a generating unit, which uses a camera, range-image sensor, and projector to detect movement and position of an operation object, acquire partial images of the document, and synthesize these images to generate a distorted-free image of the target object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Shape
If the target object is held down at both ends with hands or fingers, then the peripheral portions are flattened, but distortion in the center portion (such as folds) cannot be sufficiently suppressed
Solution Approach 1:
The patent divides the imaging process into multiple segments by capturing a series of partial images as the operation object moves across different positions. Each partial image captures a specific region where distortion is being suppressed at that moment. These segmented images are then synthesized to form a complete image with overall distortion suppression, resolving the contradiction between local flattening and global image quality.
Solution Approach 2:
The patent employs dynamic imaging by moving the operation object across the target object during the imaging process. Rather than attempting to hold down the entire document statically, the system dynamically captures multiple partial images at different positions and times, allowing distortion suppression to be applied sequentially across different regions, thereby achieving both local flattening and global recognition accuracy.
2Device complexity
If the entire target object is imaged in a single image, then the imaging process is simple, but distortion in various portions cannot be sufficiently suppressed
Solution Approach 1:
The patent segments the imaging task into multiple partial image acquisitions taken at different positions as the operation object moves. This segmentation allows distortion suppression to be applied to each specific region individually, achieving better overall shape control while maintaining a relatively simple imaging process through automated sequential capture and synthesis.
Solution Approach 2:
The patent applies partial action by capturing multiple partial images of different regions rather than attempting to capture the entire distorted object in a single image. Each partial image is acquired with local distortion suppression, and the excessive number of partial images beyond what would be needed for a single shot are synthesized to create a final image with comprehensive distortion correction.
3Shape
If multiple partial images are acquired and synthesized, then distortion suppression is improved, but the imaging time increases
Solution Approach 1:
The patent maintains continuity of useful action by continuously capturing partial images as the operation object moves across the target object. Rather than stopping and starting the imaging process, the system continuously acquires images throughout the movement, ensuring that every moment of the operation contributes to the final synthesized image, thereby minimizing idle time while achieving comprehensive distortion suppression.
Solution Approach 2:
The patent employs periodic action by acquiring partial images at regular intervals or at predetermined positions during the movement of the operation object. This periodic capture strategy ensures consistent coverage of the entire target object while maintaining a systematic and efficient imaging rhythm, balancing the need for multiple images with the constraint of imaging time.
Data Source
AI summary
Provided is an apparatus, including: an image input unit configured to receive input of an image on which a target object placed on an operation plane imaged as a subject; an operation detecting unit configured to detect movement of a predetermined operation object in a state of being in contact with the target object using the image; a partial-image acquiring unit configured to acquire a partial image corresponding to a portion of the target object from the image in accordance with a position of the predetermined operation object being moved in a case where the movement of the predetermined operation object is detected by the operation detecting unit; and a generating unit configured to generate an image representing the target object using a plurality of the partial images acquired by the partial-image acquiring unit.