halcon 检测大米个数

最新推荐文章于 2024-01-11 22:38:41 发布

原创最新推荐文章于 2024-01-11 22:38:41 发布 · 1.2k 阅读

1 ·

CC 4.0 BY-SA版权

halcon 专栏收录该内容

1 篇文章

订阅专栏

本文介绍了一种使用Halcon进行图像处理的方法，通过读取图像、阈值分割、连接区域、计数对象等步骤，实现了对稻米颗粒的有效检测和标记。详细展示了如何通过编程实现对图像中较小目标的直接标记以及对较大目标进行形态学操作后的分割标记。

read_image (Image, 'rice')
illuminate (Image, ImageIlluminate, 101, 101, 0.7)
invert_image (ImageIlluminate, ImageInvert)
set_window_param (3600, 'background_color', 'black')
binary_threshold (ImageInvert, Region, 'max_separability', 'dark', UsedThreshold)
connection (Region, ConnectedRegions)
count_obj (ConnectedRegions, Number)
area_center (ConnectedRegions, Area, Row, Column)
tuple_max (Area, Max)

disp_image (Image, 3600)
i:=1
MaxArea:=Max*0.6 
for Index := 1 to Number by 1
    select_obj (ConnectedRegions, ObjectSelected, Index)
    area_center (ObjectSelected, SmallArea, Row, Column)
    if (SmallArea <= MaxArea)
        disp_message (3600, i, 'image', Row, Column, 'black', 'false')
        i:=i+1
    else
        opening_circle (ObjectSelected, RegionOpening, 3.5)
        erosion_circle (RegionOpening, RegionErosion, 1.5)
        connection (RegionErosion, ConnectedRegions1)
        count_obj (ConnectedRegions1, Number2)
        for Index1 := 1 to Number2 by 1
            select_obj (ConnectedRegions1, ObjectSelected1, Index1)
            area_center (ObjectSelected1, Area1, Row1, Column1)
            disp_message (3600, i, 'image', Row1, Column1, 'black', 'false')
            i:=i+1
        endfor
    endif
    stop ()
endfor