Halcon extract orange (Orange)

demand ： Extract oranges from the following pictures

 

  Code ：

dev_set_draw ('margin')
dev_set_line_width (3)
read_image (Image, 'C:/Users/ASUS/Desktop/halcon/ Orange picture /citrus_fruits_01.png')
* List all the pictures in the folder 
list_files ('C:/Users/ASUS/Desktop/halcon/ Orange picture ', ['files','follow_links','recursive'], ImageFiles)
* Select the appropriate element 
tuple_regexp_select (ImageFiles, ['\\.(tif|tiff|gif|bmp|jpg|jpeg|jp2|png|pcx|pgm|ppm|pbm|xwd|ima|hobj)$','ignore_case'], ImageFiles)
dev_open_window_fit_image (Image, 0, 0, -1, -1, WindowHandle)
* Extract the oranges in each picture by cycling 
for Index := 0 to |ImageFiles|-1 by 1
    read_image (Image1, ImageFiles[Index])
    rgb1_to_gray (Image1, GrayImage)
    decompose3 (Image1, ImageR, ImageG, ImageB)
    threshold (ImageR, Regions, 40, 255)
    connection (Regions, ConnectedRegions)
    select_shape (ConnectedRegions, SelectedRegions, ['circularity','area'], 'and', [0.8,150], [1,99999])
    * display picture 
     dev_display(Image1)
    * Show the selected area 
    dev_display(SelectedRegions)
endfor

Effect display ：

  explain

        1. Why in the red passage （ Left ） Perform threshold segmentation on ？ Because the background of the picture is indigo , If directly in the original image to gray image （ Right ） Upper threshold segmentation , The effect of some pictures is not good , As shown in the figure below ：

                                                             

        2. Picture material ：halcon Orange extract picture material - Internet document resources -CSDN download