Halcon image calibration enables subsequent image processing to become the same as the template image

Choose any picture

After the rotation correction of this picture , The image becomes a template image . Its area region The location is as shown in the figure ：

 

 

When a new picture comes , Let it relate to region Compare , There is obvious deviation from the position of the formwork , As shown in the figure ：

Our goal is to make this new image coincide with the template area after processing . As shown in the figure below ：

In fact, the overall function is similar to rotating matting , Only the functions used are a little different .

 

 

Code aggregation ：

The image used in the code is ：C:\Users\HJ\AppData\Roaming\MVTec\HALCON-21.05-Progress\examples\images\blister（ Install it on your own halcon To find ）

* Read a picture for processing , And take the processed figure as a template , Let the following pictures look like this one 
read_image (Image, 'C:/Users/HJ/Desktop/test_image/blister/blister_01.png')
threshold(Image, Region, 100, 255)
connection(Region, ConnectedRegions)
select_shape(ConnectedRegions, SelectedRegions, 'area', 'and', 15000, 9999999)
shape_trans(SelectedRegions, RegionTrans, 'convex')
* Calculation region The direction of （ angle ）,Phi Is the farthest point of the area and x Angle between axes 
orientation_region(RegionTrans, Phi)
* Get the center of the area 
area_center(RegionTrans, Area, Row, Column)
if ((1.57<Phi and Phi<3.142) or (-3.142<Phi and Phi<-1.57))
    vector_angle_to_rigid(Row, Column, Phi, Row, Column, 3.14159, HomMat2D)
else
    vector_angle_to_rigid(Row, Column, Phi, Row, Column, 0, HomMat2D)
endif
* Rotate pictures and areas 
affine_trans_image(Image, ImageAffineTrans, HomMat2D, 'constant', 'false')
affine_trans_region(RegionTrans, RegionTrans, HomMat2D, 'nearest_neighbor')
* Calculate the center of the area after rotation ,  That is, the center of the template area 
area_center (RegionTrans, AreaRef, RowRef, ColumnRef)


list_files ('C:/Users/HJ/Desktop/test_image/blister', ['files','follow_links'], ImageFiles)
tuple_regexp_select (ImageFiles, ['\\.(tif|tiff|gif|bmp|jpg|jpeg|jp2|png|pcx|pgm|ppm|pbm|xwd|ima|hobj)$','ignore_case'], ImageFiles)
Width := 550
Height := 400
for Index := 0 to |ImageFiles| - 1 by 1
    * Data preprocessing 
    read_image (Image2, ImageFiles[Index])
    threshold(Image2, Region_Each, 100, 255)
    connection(Region_Each, ConnectedRegions_Each)
    select_shape(ConnectedRegions_Each, SelectedRegions_Each, 'area', 'and', 15000, 9999999)
    shape_trans(SelectedRegions_Each, RegionTrans_Each, 'convex')
    
    * Calculate the direction of the current area （ angle ）
    orientation_region(RegionTrans_Each, Phi_Each)
    * Get the center of the current area 
    area_center(RegionTrans_Each, Area_Each, Row_Each, Column_Each)
    
    if ((1.57<Phi_Each and Phi_Each<3.142) or (-3.142<Phi_Each and Phi_Each<-1.57))
        * Use the center of the current area 、 angle      and    Calculate the center and angle of the template area ,  Generate a matrix of affine transformation 
        vector_angle_to_rigid(Row_Each, Column_Each, Phi_Each, RowRef, ColumnRef, 3.14159, HomMat2D1)
    else
        vector_angle_to_rigid(Row_Each, Column_Each, Phi_Each, RowRef, ColumnRef, 0, HomMat2D1)
    endif
    * Adjust the picture according to the affine transformation matrix , Become the same as the template image 
    affine_trans_image(Image2, ImageAffineTrans_Each, HomMat2D1, 'constant', 'false')
endfor