当前位置:   article > 正文

书接上文,基于藏文手写数字数据开发构建yolov5n轻量级藏文手写数字检测识别系统_yolov5手写体印刷体分类检测

yolov5手写体印刷体分类检测

在上一篇文章中:
《python基于轻量级CNN模型开发构建手写藏文数字识别系统》

开发实现了轻量级的藏文手写数字识别系统,这里主要是想基于前文的数据,整合目标检测模型来进一步挖掘藏文手写数字数据集的可玩性,基于yolov5n开发构建轻量级的藏文手写数字检测识别系统,首先来看效果图:

 共仿真生成了3000的样本数据集,接下来简单看下:

 YOLO格式标注文件如下所示:

 实例标注内容如下:

 VOC格式标注文件如下所示:

 实例标注内容如下所示:

  1. <annotation>
  2. <folder>JiaGuWen</folder>
  3. <filename>JPEGImages/0a48304e-c797-4686-9c2a-09eeb029404d.jpg</filename>
  4. <source>
  5. <database>The JiaGuWen Database</database>
  6. <annotation>JiaGuWen</annotation>
  7. <image>JiaGuWen</image>
  8. </source>
  9. <owner>
  10. <name>CGB</name>
  11. </owner>
  12. <size>
  13. <width>640</width>
  14. <height>640</height>
  15. <depth>3</depth>
  16. </size>
  17. <segmented>0</segmented>
  18. <object>
  19. <name>0</name>
  20. <pose>Unspecified</pose>
  21. <truncated>0</truncated>
  22. <difficult>0</difficult>
  23. <bndbox>
  24. <xmin>590</xmin>
  25. <ymin>14</ymin>
  26. <xmax>618</xmax>
  27. <ymax>42</ymax>
  28. </bndbox>
  29. </object>
  30. <object>
  31. <name>7</name>
  32. <pose>Unspecified</pose>
  33. <truncated>0</truncated>
  34. <difficult>0</difficult>
  35. <bndbox>
  36. <xmin>392</xmin>
  37. <ymin>98</ymin>
  38. <xmax>448</xmax>
  39. <ymax>154</ymax>
  40. </bndbox>
  41. </object>
  42. <object>
  43. <name>1</name>
  44. <pose>Unspecified</pose>
  45. <truncated>0</truncated>
  46. <difficult>0</difficult>
  47. <bndbox>
  48. <xmin>145</xmin>
  49. <ymin>134</ymin>
  50. <xmax>187</xmax>
  51. <ymax>176</ymax>
  52. </bndbox>
  53. </object>
  54. <object>
  55. <name>1</name>
  56. <pose>Unspecified</pose>
  57. <truncated>0</truncated>
  58. <difficult>0</difficult>
  59. <bndbox>
  60. <xmin>380</xmin>
  61. <ymin>352</ymin>
  62. <xmax>408</xmax>
  63. <ymax>380</ymax>
  64. </bndbox>
  65. </object>
  66. </annotation>

模型文件如下:

  1. # YOLOv5 本文内容由网友自发贡献,转载请注明出处:https://www.wpsshop.cn/w/小蓝xlanll/article/detail/218672
    推荐阅读
    相关标签