Shape template matching based on Halcon learning [v] find_ cocoa_ packages_ max_ deformation. Hdev routine

* This sample program demonstrates how to use shape based matching to find slightly deformed objects , Using parameter “ Maximum deformation ”;

* Please note that , To find deformed objects , Shape based matching can be applied using parameters “ Maximum deformation ” or - Apply locally deformable matching ;

* In this case , The task is to use shape based matching and parameters “ Maximum deformation ”. You can compare it with the example “ Find a deformable cocoa bag ”, Which uses local deformation matching to solve the same task . Shape based matching is significantly faster , But local deformable matching is more robust to different types of images

dev_update_off ()
* Read images 
read_image (ModelImage, 'food/cocoa_package_model')
* close window 
dev_close_window ()
* Open window adaptive image 
dev_open_window_fit_image (ModelImage, 0, 0, -1, -1, WindowHandle)
* Set display font 
set_display_font (WindowHandle, 16, 'mono', 'true', 'false')

   First step ： Create a template

* Create and display shape models 
create_shape_model (ModelImage, 'auto', rad(-20), rad(40), 'auto', 'auto', 'use_polarity', [40,60,'auto_min_size'], 10, ModelID)

* Get the outline of the template 
* notes ： The position of the template will return to the origin 
get_shape_model_contours (ModelContours, ModelID, 1)

* Get the row and column coordinates of the region 
area_center (ModelImage, Area, Row, Column)

* Affine transformation 
hom_mat2d_identity (HomMat2DIdentity)
hom_mat2d_translate (HomMat2DIdentity, Row, Column, HomMat2DTranslate)
affine_trans_contour_xld (ModelContours, ContoursAffineTrans, HomMat2DTranslate)

* Set a series of parameters 
dev_set_line_width (2)
dev_set_color ('yellow')
dev_display (ModelImage)
dev_display (ContoursAffineTrans)

disp_message (WindowHandle, 'Model image and contours', 'window', 12, 12, 'black', 'true')
disp_continue_message (WindowHandle, 'black', 'true')
stop ()

 2、 Start template matching

* The picture has 13 Zhang 
NumImages := 13
* To begin for loop 
for Index := 1 to NumImages by 1
    * Read the picture 
    read_image (Image, 'food/cocoa_packages_' + Index$'02')

    * Reduce image resolution to improve speed 
    dev_resize_window_fit_image (Image, 0, 0, -1, -1)
    * display picture     
    dev_display (Image)
    
    disp_message (WindowHandle, 'Search...', 'window', 12, 12, 'black', 'true')
    * Time before looking for template 
    count_seconds (S1)

    * Find the deformation model in the search image 
    * And display the results 

    * Start looking for templates 
    find_shape_model (Image, ModelID, rad(-20), rad(40), 0.6, 0, 0.5, ['least_squares','max_deformation 16'], 0, 0.4, Row, Column, Angle, Score)
    * Time after finding the template 
    count_seconds (S2)
    Time := (S2 - S1) * 1000

    * Set the color 
    dev_set_color ('green')
    * Display the matching results 
    dev_display_shape_matching_results (ModelID, 'green', Row, Column, Angle, 1, 1, ModelID)
    disp_message (WindowHandle, |Score| + ' matches found in ' + Time$'3.1f' + ' ms', 'window', 12, 12, 'black', 'true')

    * Show the results below the picture 
    disp_message (WindowHandle, 'Score: ' + Score$'.2f', 'image', 350, Column - 80, 'black', 'true')

    * If the index is smaller than the number of pictures , Continued to 
    if (Index < NumImages)
        disp_continue_message (WindowHandle, 'black', 'true')
        stop ()
    endif
endfor

 3、 Clear template

clear_shape_model (ModelID)

---

---

  notes ： Here's how to find template parameters

* Start looking for templates 
    find_shape_model (Image, ModelID, rad(-20), rad(40), 0.6, 0, 0.5, ['least_squares','max_deformation 16'], 0, 0.4, Row, Column, Angle, Score)

 1、 In some applications , The accuracy requirement is very high . In these cases , The attitude of the model can be determined by least square adjustment . And “interpolation” contrary , This mode requires additional computation time . Different least squares methods can be used to adjust the mode （'least_squares','least_squares_high' and 'least_squares_very_high'） To determine the accuracy of the minimum distance being searched . However , The higher the selection accuracy , The longer the sub-pixel extraction will take . Usually ,SubPixel Should be set to “interpolation”. If the least square adjustment is needed , You should choose “least_squares”, Because of this Will result in the best trade-off between runtime and accuracy .

2、 An object that is slightly deformed relative to the model , In some cases , Cannot be found or even if found , Accuracy is also very low . For such objects , It can be in the parameter SubPixel Maximum allowable object deformation for medium and extra frontal delivery . Deformation must be specified in pixels . This can be done by passing optional parameter values “max_deformation”, Followed by a 0 To 32 The integer value between （ In the same string ）, This value specifies the maximum deformation . for example , If the shape of the object can deform as much as... Relative to the shape stored in the model 2 Pixel , Must be in SubPixel Pass values in “max_deformation 2”. Transfer value “max_deformation 0” when , When searching, the object must not be deformed , And not set “max_deformation” It does the same thing .

3、 Please note that , When the maximum deformation value is set high, it usually leads to an increase in running time . Besides , Select the higher the deformation value , The higher the risk of discovering instances of the wrong model . These two problems mainly appear when searching for small objects or objects with fine structures . This is because this kind of object will lose its characteristic shape when it is highly deformed , And feature shape is very important for powerful search . Also pay attention to , For higher deformations , If there is clutter near the object , The accuracy of partially occluded objects may be reduced . therefore , The maximum deformation should be as small as possible , Set it higher only when necessary .