一、待识别字符的图像：

二、运行环境：

Halcon-12.0

三、代码如下：

read_image (Image, 'C:/Users/研发/Desktop/捕获.PNG')

*缩小定义域

draw_rectangle1 (3600, Row1, Column1, Row2, Column2)

gen_rectangle1 (Rectangle, Row1, Column1, Row2, Column2)

reduce_domain (Image, Rectangle, Image)

*增强孤立点或团状缺陷，金属上雕刻字符的图像，需要此算子处理后，才能容易的分割出来字符

gray_range_rect (Image, ImageResult, 7, 7)

*图像取反，字符识别特定是识别亮背景下的暗字符

invert_image (ImageResult, ImageResult)

threshold (ImageResult, Region, 0, 140)

*闭运算，填补细小孔洞

closing_rectangle1 (Region, RegionClosing, 2, 2)

connection (RegionClosing, ConnectedRegions)

select_shape (ConnectedRegions, SelectedRegions, 'area', 'and', 500, 99999)

*将区域按照自定义大小分开，40,100表示字符大小（矩形）

partition_rectangle (SelectedRegions, Partitioned, 40,100)

*smallest_rectangle1 (Partitioned, Row11, Column11, Row21, Column21)

*gen_rectangle1 (Rectangle1, Row11, Column11, Row21, Column21)

sort_region (Partitioned, SortedRegions, 'first_point', 'true', 'column')

*根据字符类型选择合适的文件，也可以自己训练文件（可参照本人另一篇博客：Halcon字符识别）

read_ocr_class_mlp ('Industrial_0-9A-Z_NoRej.omc', OCRHandle)

count_obj (SortedRegions, Number)

*循环识别单个字符

for Index:=1 to Number by 1

    select_obj (SortedRegions, ObjectSelected, Index)

    do_ocr_multi_class_mlp (ObjectSelected, ImageResult,OCRHandle, Class, Confidence)   

endfor

clear_ocr_class_mlp (OCRHandle)