Template matching alignment method based on affine iterative transformation
An iterative transformation and template matching technology, applied in the field of template matching and alignment based on affine iterative transformation, can solve problems such as unsatisfied, deformed, unsatisfied image alignment processing, etc., and achieve the effect of improving efficiency and accuracy
Active Publication Date: 2020-07-28
南京红松信息技术有限公司
View PDF4 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004]
During the actual use of automated review products, due to the influence of many reasons such as printing and scanning, the image has certain distortions, deformations, scaling, etc.
In order to solve such practical problems in images, image preprocessing will become urgently needed, and simple translation and scaling can no longer satisfy the commonly used image alignment processing. Although affine transformation is an effective image correction and transformation method, But there are also unsatisfactory situations when dealing with image alignment tasks
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0028] Example: such as figure 1 As shown, the template matching alignment method based on affine iterative transformation specifically includes the following steps:
[0029] S1 Approximate Alignment: Firstly, the scanned test paper of a test paper is divided into two pieces, including the main paper and the sub-paper, and the images of the main paper and the sub-paper are obtained, and then the character coordinates of the images of the main paper and the sub-paper are respectively located to obtain the characters of the main paper coordinates and character coordinates of the sub-volume, moving the sub-volume so that the character coordinates of the parent volume and the character coordinates of the sub-volume are aligned;
[0030] The step S1 also includes the calculated offset and scaling s of the character coordinates of the parent volume and the character coordinates of the sub-volume, and then moves the sub-volume according to the offset, so that the The character coord...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a template matching alignment method based on affine iterative transformation, which comprises the following steps of: S1, setting a scanning test paper of a test paper to comprise a primary volume and a secondary volume, respectively positioning character coordinates of a primary volume image and a secondary volume image, and moving the secondary volume to enable the character coordinates of the primary volume and the character coordinates of the secondary volume to reach an alignment state; s2, positioning coordinates of approximately aligned sub-volume characters, acquiring an X axis and a Y axis, acquiring coordinates of a first character at the upper left corner, taking the coordinates of the first character as a starting point, and cutting the starting pointto obtain a to-be-corrected image; s3, converting the to-be-corrected image to the corrected image matrix through affine iteration to obtain a corrected image; s4, performing template matching on thelocal correction image after affine transformation and the local template image once for each affine iteration, performing continuous iteration, and stopping affine iteration until matching succeeds,thereby obtaining an affine transformation matrix; s5, applying the affine transformation matrix to transformation in the sub-volume to complete alignment of the mother volume and the sub-volume; andthe alignment efficiency and accuracy are improved.
Description
technical field [0001] The invention relates to the technical field of text image processing, in particular to a template matching alignment method based on affine iterative transformation. Background technique [0002] As an important branch of artificial intelligence, machine vision has achieved rapid development in recent years and has been widely used in various industries, greatly improving human productivity and production methods, not only reducing people's labor, but also easing It relieves people's work burden and brings great convenience and improvement to people's life style. [0003] In the field of education, computer vision technology has been fully applied to the automated marking system, which can help teachers to automate the marking of test papers. This not only greatly reduces the time used by teachers to manually correct test papers, but also improves the work efficiency of teachers. Today's computer vision technology has made brilliant achievements, an...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06K9/00G06K9/62G06T3/40
CPCG06T3/4084G06V30/40G06V30/43G06V10/751Y02P90/30
Inventor 田博帆
Owner 南京红松信息技术有限公司