Optical Code Sensor With ML Motion Blur Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical sensors struggle to reliably detect codes moving at relative speeds due to motion blur, which blurs the contrast patterns and prevents effective decoding, limiting their application possibilities.
Innovation Solution
Employ a machine learning model with neural networks trained to determine and compensate for motion blur in captured images using a learning phase with input data, allowing for reliable decoding even at varying speeds and conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If codes move at higher relative speeds to the optical sensor, then productivity increases, but motion blur increases causing contrast patterns to become indistinct and code decoding to fail
Solution Approach 1:
The patent applies preliminary action by capturing multiple images at different exposure times before the code fully passes through the field of view. The evaluation unit processes these images in real-time to detect the contrast pattern and determine code speed, then uses this information to adjust exposure settings for subsequent images. This proactive approach allows the system to maintain reliable decoding accuracy even when codes move at higher relative speeds, effectively resolving the contradiction between productivity and reliability.
2Manufacturing precision
If traditional filters are used to reduce motion blur, then some blur reduction is achieved, but image artifacts are introduced that make code recognition difficult or impossible
Solution Approach 1:
The patent replaces traditional mechanical/optical filtering methods with a computational approach. Instead of using physical filters that manipulate light paths and introduce artifacts, the evaluation unit uses image processing algorithms to selectively enhance sharp edges and transitions in the captured images. This substitution of mechanical filtering with computational processing achieves motion blur reduction while preserving image fidelity and avoiding harmful artifacts, thereby resolving the contradiction between image sharpness and artifact generation.
3Measurement precision
If exposure time is increased to improve image quality, then contrast patterns become sharper, but motion blur increases for moving codes
Solution Approach 1:
The patent applies dynamics by making the exposure time variable rather than fixed. The evaluation unit continuously monitors the detected code speed and dynamically adjusts the exposure time parameter accordingly. When codes move slowly, longer exposure times are used to maximize contrast pattern sharpness. When codes move quickly, shorter exposure times prevent excessive motion blur. This dynamic adaptation allows the system to maintain optimal measurement precision across a wide range of relative speeds, resolving the contradiction between contrast sharpness and speed tolerance.
Data Source
Figure 1~2
Figure 3~4b
Figure 5~8
AI summary
The invention relates to an optical sensor (1) for detecting codes (3) with an image sensor (12), by means of an optical sensor (1) by means of which images of codes (3) are recorded, and with an evaluation unit (15) having a decoder (17). Codes (3) detected by the image sensor (12) are decoded in the decoder (17). The evaluation unit (15) has a machine learning model (16) with at least one neural network (18, 18a, 18.1, ... to 18.k). The neural network (18, 18a, 18.1, ... or 18.k) is trained with training data in a learning mode. With the machine learning model (16), a motion blur in an image of the code (3) caused by a relative movement of a detected code (3) relative to the optical sensor (1) is determined according to magnitude (u) and phase (φ) as an intermediate quantity.Using this intermediate value, the motion blur contained in the code image (3) is compensated for by deconvolution in the machine learning model (16). The resulting processed image of the code (3) is fed to the decoder (17).