TDI Image Sensor Speed Adaptation for Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional TDI methods often result in blurred and artifact-ridden images due to variations in object transport speed and non-flat object surfaces, leading to inaccuracies in image recording.
Innovation Solution
A method using a multiline CMOS color sensor to create sharp, artifact-free images by identifying subsets of matrix data structure pixels with specific distance and brightness criteria, synchronizing movement speed, and determining object distance through angle calculations, allowing for precise color and brightness determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional TDI methods are used to record moving objects, then recording speed can be maintained, but image quality deteriorates due to blurring and movement artifacts when transport speed varies
Solution Approach 1:
The patent implements dynamic adaptation of the TDI readout timing to match the actual transport speed of the object. The system continuously adjusts the synchronization between sensor readout and object movement, allowing the recording system to maintain optimal image quality across varying transport speeds without sacrificing recording throughput
Solution Approach 2:
The system incorporates feedback mechanisms that monitor transport speed variations and adjust the TDI readout timing accordingly. This closed-loop control ensures that the image integration timing remains synchronized with the actual object position, preventing blurring and artifacts while maintaining high recording speeds
2Adaptability or versatility
If conventional TDI methods are used for non-flat objects, then all areas can be recorded, but measurement precision deteriorates due to varying distances from the image recording unit
Solution Approach 1:
The patent applies local quality adjustment by individually processing pixel data based on their specific distance characteristics. Pixels from different depth planes are identified and processed with appropriate parameters, allowing accurate measurement and imaging of both flat and non-flat surfaces simultaneously without compromising precision for any specific region
3Measurement precision
If subsets of matrix data structure are searched with strict criteria to determine object distance, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent segments the matrix data structure into distinct subsets based on spatial and intensity characteristics. By dividing the data processing into manageable segments with specific search criteria, the system achieves precise distance determination through structured analysis rather than exhaustive processing of all pixels, thereby improving measurement precision while controlling computational complexity
Data Source
Figure 1a~1b
Figure 1c
Figure 2a
AI summary
1. Method for creating an image of an object (1) by means of an image acquisition unit (10) comprising an area sensor (11), - wherein the area sensor (11) has a number of sensor pixels (14) arranged in a grid in rows (12) and columns (13), each of which is assigned a row index (iz) and a column index (is) according to its arrangement, - wherein the object (20) and the image acquisition unit (10) perform a translational relative movement parallel to the direction of the columns (13) of the area sensor (11), - wherein the object (20) is moved on an object plane (21) located at a predetermined distance (d) from the image acquisition unit (10) relative to the image acquisition unit (10) and the recording area of the image acquisition unit (10), and - wherein the color and brightness measurement values created by each column (13) of sensor pixels (14) of the image acquisition unit (10) are each displayed in an image-like,a two-dimensional matrix data structure Mx(y, t) with two indices (y, t) and two coordinate directions are provided, wherein the first index (y) and the first coordinate (y) correspond to the row position of the sensor pixel (14) that creates the color and brightness measurement value, and the second index (t) and the second coordinate (t) correspond to the recording time, and the column index (x) denotes the position of the column (13), - wherein a two-dimensional image data structure (B) is created as an image, the number of rows of which corresponds to the number of image sensors (14) per row (12) and the number of columns of which is determined by the number of recordings made, wherein the image data structure (B) has a first index (x) that corresponds to the image column (13) that records the color and/or brightness measurement value and has a second index (t) that corresponds to the respective recording time and the relative movement of the object (1), wherein for at least one,in particular, each of the image columns (13) is searched separately – within the respective matrix data structure Mx(y, t) assigned to the image column (13) with the column index (x) – for contiguous, especially linear, image sections (A) with the same or similar color that run at an angle to the first and second coordinate directions; – that separately for each of the individual image sections (A) – the sum or a mean value (S) of the color and brightness measurements of the pixels belonging to the image section (A) is determined; – a second index or time point (t) is determined at which the image section (A) intersects a predefined straight line in the image of the matrix data structure (M) running parallel to the second coordinate direction; and – the sum or mean value (S) is stored and kept available in the image data structure (B) at the position designated by the column index (x) and by the second index or time point (t).