数独-识别已定位的数字

最新推荐文章于 2024-09-16 14:02:00 发布

qingye2008

最新推荐文章于 2024-09-16 14:02:00 发布

阅读量1.4k

点赞数 1

CC 4.0 BY-SA版权

分类专栏： python tensorflow

本文链接：https://blog.youkuaiyun.com/qingye2008/article/details/79085267

python 同时被 2 个专栏收录

2 篇文章

订阅专栏

tensorflow

2 篇文章

订阅专栏

该博客旨在介绍如何在已定位的数独图片中，通过TensorFlow技术进行数字内容的切割与识别，最终将识别结果构建成numpy数组。文章提到训练准确率曲线，并预告后续会分享相关代码。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

目的: 图片已经定位,需要切割每个有内容的单元格,使用TensorFlow识别数字内容, 构造numpy的数组

训练准确率曲线

有效轮廓 valid contours lenth is :26
26
分割图片耗时: 0.1448
2018-01-17 14:55:24.636066: W tensorflow/core/framework/op_def_util.cc:334] Op BatchNormWithGlobalNormalization is deprecated. It will cease to work in GraphDef version 9. Use tf.nn.batch_normalization().
fileName:/tmp/upload/1516172123.001_Wait-4.png___0_1, result:6, score:0.864061 
fileName:/tmp/upload/1516172123.001_Wait-4.png___0_3, result:9, score:0.851601 
fileName:/tmp/upload/1516172123.001_Wait-4.png___0_5, result:4, score:0.975630 
fileName:/tmp/upload/1516172123.001_Wait-4.png___0_6, result:5, score:0.850700 
fileName:/tmp/upload/1516172123.001_Wait-4.png___0_7, result:8, score:0.983153 
fileName:/tmp/upload/1516172123.001_Wait-4.png___1_4, result:2, score:0.963580 
fileName:/tmp/upload/1516172123.001_Wait-4.png___1_6, result:3, score:0.796880 
fileName:/tmp/upload/1516172123.001_Wait-4.png___1_8, result:4, score:0.933991 
fileName:/tmp/upload/1516172123.001_Wait-4.png___2_1, result:4, score:0.943660 
fileName:/tmp/upload/1516172123.001_Wait-4.png___2_2, result:1, score:0.762967 
fileName:/tmp/upload/1516172123.001_Wait-4.png___3_0, result:4, score:0.882343 
fileName:/tmp/upload/1516172123.001_Wait-4.png___3_5, result:5, score:0.775673 
fileName:/tmp/upload/1516172123.001_Wait-4.png___3_6, result:2, score:0.944187 
fileName:/tmp/upload/1516172123.001_Wait-4.png___5_2, result:6, score:0.943249 
fileName:/tmp/upload/1516172123.001_Wait-4.png___5_3, result:4, score:0.986753 
fileName:/tmp/upload/1516172123.001_Wait-4.png___5_8, result:7, score:0.986366 
fileName:/tmp/upload/1516172123.001_Wait-4.png___6_6, result:4, score:0.988631 
fileName:/tmp/upload/1516172123.001_Wait-4.png___6_7, result:7, score:0.991990 
fileName:/tmp/upload/1516172123.001_Wait-4.png___7_0, result:5, score:0.886496 
fileName:/tmp/upload/1516172123.001_Wait-4.png___7_2, result:4, score:0.976941 
fileName:/tmp/upload/1516172123.001_Wait-4.png___7_4, result:9, score:0.861926 
fileName:/tmp/upload/1516172123.001_Wait-4.png___8_1, result:7, score:0.914692 
fileName:/tmp/upload/1516172123.001_Wait-4.png___8_2, result:9, score:0.733868 
Warning**** fileName:/tmp/upload/1516172123.001_Wait-4.png (8,3) result:1, score:0.648592 
fileName:/tmp/upload/1516172123.001_Wait-4.png___8_5, result:8, score:0.989103 
fileName:/tmp/upload/1516172123.001_Wait-4.png___8_7, result:5, score:0.783846 
识别图片
[[0 0 0 4 0 0 0 5 0]
 [6 0 4 0 0 0 0 0 7]
 [0 0 1 0 0 6 0 4 9]
 [9 0 0 0 0 4 0 0 1]
 [0 2 0 0 0 0 0 9 0]
 [4 0 0 5 0 0 0 0 8]
 [5 3 0 2 0 0 4 0 0]
 [8 0 0 0 0 0 7 0 5]
 [0 4 0 0 0 7 0 0 0]]

稍后更新代码....