python cnn tensorflow 车牌识别 网络模型

Posted

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了python cnn tensorflow 车牌识别 网络模型相关的知识,希望对你有一定的参考价值。

1、模型结构图

 

2、随机测试模型

 

 

 

 

 

 

 3、训练logs

2020-05-10T11:28:20.491640: Step 4, loss_total = 28.22, acc = 2.23%, sec/batch = 1.23
2020-05-10T11:28:27.849279: Step 9, loss_total = 26.00, acc = 2.23%, sec/batch = 1.23
2020-05-10T11:28:35.128835: Step 14, loss_total = 25.52, acc = 2.23%, sec/batch = 1.23
2020-05-10T11:28:42.404306: Step 19, loss_total = 25.57, acc = 3.12%, sec/batch = 1.23
2020-05-10T11:28:49.699494: Step 24, loss_total = 24.92, acc = 2.23%, sec/batch = 1.23
2020-05-10T11:28:57.041452: Step 29, loss_total = 24.89, acc = 2.68%, sec/batch = 1.25
2020-05-10T11:29:04.305255: Step 34, loss_total = 24.91, acc = 1.79%, sec/batch = 1.23
2020-05-10T11:29:11.584816: Step 39, loss_total = 24.80, acc = 2.68%, sec/batch = 1.23
2020-05-10T11:29:18.879945: Step 44, loss_total = 24.45, acc = 4.02%, sec/batch = 1.23
2020-05-10T11:29:26.168018: Step 49, loss_total = 24.64, acc = 1.79%, sec/batch = 1.23
2020-05-10T11:29:33.478765: Step 54, loss_total = 24.76, acc = 1.79%, sec/batch = 1.23
2020-05-10T11:29:40.758330: Step 59, loss_total = 24.61, acc = 2.68%, sec/batch = 1.23
2020-05-10T11:29:48.023288: Step 64, loss_total = 24.36, acc = 4.91%, sec/batch = 1.23
2020-05-10T11:29:55.302794: Step 69, loss_total = 24.48, acc = 1.79%, sec/batch = 1.23
2020-05-10T11:30:02.613572: Step 74, loss_total = 24.69, acc = 2.23%, sec/batch = 1.23
2020-05-10T11:30:09.940001: Step 79, loss_total = 24.43, acc = 4.02%, sec/batch = 1.25
2020-05-10T11:30:17.219536: Step 84, loss_total = 24.31, acc = 6.25%, sec/batch = 1.23
2020-05-10T11:30:24.499079: Step 89, loss_total = 24.52, acc = 3.12%, sec/batch = 1.23
2020-05-10T11:30:31.778604: Step 94, loss_total = 24.50, acc = 3.12%, sec/batch = 1.23
2020-05-10T11:30:39.081043: Step 99, loss_total = 24.64, acc = 3.12%, sec/batch = 1.23
2020-05-10T11:30:46.423060: Step 104, loss_total = 24.22, acc = 4.02%, sec/batch = 1.23
2020-05-10T11:30:53.733840: Step 109, loss_total = 24.32, acc = 4.02%, sec/batch = 1.23
2020-05-10T11:31:01.048162: Step 114, loss_total = 24.24, acc = 3.57%, sec/batch = 1.25
2020-05-10T11:31:08.327694: Step 119, loss_total = 24.56, acc = 3.12%, sec/batch = 1.23
2020-05-10T11:31:15.607261: Step 124, loss_total = 24.18, acc = 3.57%, sec/batch = 1.23
2020-05-10T11:31:22.871144: Step 129, loss_total = 24.02, acc = 5.36%, sec/batch = 1.25
2020-05-10T11:31:30.150680: Step 134, loss_total = 24.25, acc = 1.34%, sec/batch = 1.23
2020-05-10T11:31:37.555186: Step 139, loss_total = 23.99, acc = 4.46%, sec/batch = 1.23
2020-05-10T11:31:44.975348: Step 144, loss_total = 23.95, acc = 6.25%, sec/batch = 1.22
2020-05-10T11:31:52.290728: Step 149, loss_total = 23.90, acc = 6.70%, sec/batch = 1.25
2020-05-10T11:31:59.632786: Step 154, loss_total = 23.57, acc = 7.14%, sec/batch = 1.23
2020-05-10T11:32:06.959193: Step 159, loss_total = 23.50, acc = 7.59%, sec/batch = 1.25
2020-05-10T11:32:14.348032: Step 164, loss_total = 23.57, acc = 8.04%, sec/batch = 1.25
2020-05-10T11:32:21.658839: Step 169, loss_total = 23.37, acc = 8.04%, sec/batch = 1.23
2020-05-10T11:32:28.969618: Step 174, loss_total = 23.02, acc = 9.38%, sec/batch = 1.23
2020-05-10T11:32:36.280367: Step 179, loss_total = 22.52, acc = 11.61%, sec/batch = 1.23
2020-05-10T11:32:43.638034: Step 184, loss_total = 22.14, acc = 17.41%, sec/batch = 1.23
2020-05-10T11:32:50.917574: Step 189, loss_total = 21.72, acc = 14.29%, sec/batch = 1.23
2020-05-10T11:32:58.197109: Step 194, loss_total = 21.08, acc = 16.96%, sec/batch = 1.25
2020-05-10T11:33:05.492796: Step 199, loss_total = 19.85, acc = 20.54%, sec/batch = 1.23
2020-05-10T11:33:12.756712: Step 204, loss_total = 19.31, acc = 20.98%, sec/batch = 1.23
2020-05-10T11:33:20.036247: Step 209, loss_total = 18.58, acc = 25.45%, sec/batch = 1.25
2020-05-10T11:33:27.346995: Step 214, loss_total = 17.98, acc = 26.34%, sec/batch = 1.23
2020-05-10T11:33:34.642180: Step 219, loss_total = 16.82, acc = 29.91%, sec/batch = 1.23
2020-05-10T11:33:41.925727: Step 224, loss_total = 15.64, acc = 33.48%, sec/batch = 1.23
2020-05-10T11:33:49.613429: Step 229, loss_total = 14.63, acc = 35.27%, sec/batch = 1.25
2020-05-10T11:33:57.418544: Step 234, loss_total = 15.99, acc = 34.38%, sec/batch = 1.23
2020-05-10T11:34:04.807462: Step 239, loss_total = 14.19, acc = 38.84%, sec/batch = 1.23
2020-05-10T11:34:12.213516: Step 244, loss_total = 11.39, acc = 50.45%, sec/batch = 1.23
2020-05-10T11:34:19.539916: Step 249, loss_total = 10.45, acc = 57.14%, sec/batch = 1.23
2020-05-10T11:34:26.849437: Step 254, loss_total = 12.10, acc = 50.00%, sec/batch = 1.22
2020-05-10T11:34:34.338093: Step 259, loss_total = 11.08, acc = 53.12%, sec/batch = 1.23
2020-05-10T11:34:41.779356: Step 264, loss_total = 10.83, acc = 54.02%, sec/batch = 1.25
2020-05-10T11:34:49.126978: Step 269, loss_total = 9.90, acc = 56.25%, sec/batch = 1.23
2020-05-10T11:34:56.618579: Step 274, loss_total = 8.89, acc = 62.05%, sec/batch = 1.25
2020-05-10T11:35:04.464056: Step 279, loss_total = 9.99, acc = 57.14%, sec/batch = 1.23
2020-05-10T11:35:12.009148: Step 284, loss_total = 7.53, acc = 62.50%, sec/batch = 1.23
2020-05-10T11:35:19.369306: Step 289, loss_total = 7.50, acc = 68.75%, sec/batch = 1.23
2020-05-10T11:35:26.817130: Step 294, loss_total = 7.53, acc = 71.43%, sec/batch = 1.23
2020-05-10T11:35:34.675697: Step 299, loss_total = 7.39, acc = 66.52%, sec/batch = 1.25
2020-05-10T11:35:42.111412: Step 304, loss_total = 6.49, acc = 69.64%, sec/batch = 1.23
2020-05-10T11:35:49.533588: Step 309, loss_total = 8.32, acc = 63.39%, sec/batch = 1.23
2020-05-10T11:35:57.404685: Step 314, loss_total = 7.38, acc = 68.30%, sec/batch = 1.23
2020-05-10T11:36:04.730550: Step 319, loss_total = 7.16, acc = 66.96%, sec/batch = 1.23
2020-05-10T11:36:12.088189: Step 324, loss_total = 7.15, acc = 69.64%, sec/batch = 1.23
2020-05-10T11:36:19.955799: Step 329, loss_total = 6.30, acc = 73.21%, sec/batch = 1.24
2020-05-10T11:36:27.932717: Step 334, loss_total = 4.79, acc = 80.36%, sec/batch = 1.25
2020-05-10T11:36:35.384052: Step 339, loss_total = 6.00, acc = 76.34%, sec/batch = 1.25
2020-05-10T11:36:43.148386: Step 344, loss_total = 5.83, acc = 76.34%, sec/batch = 1.25
2020-05-10T11:36:50.667842: Step 349, loss_total = 5.17, acc = 78.57%, sec/batch = 1.23
2020-05-10T11:36:57.956958: Step 354, loss_total = 4.72, acc = 78.57%, sec/batch = 1.23
2020-05-10T11:37:05.848736: Step 359, loss_total = 4.90, acc = 79.46%, sec/batch = 1.25
2020-05-10T11:37:13.894292: Step 364, loss_total = 4.81, acc = 79.91%, sec/batch = 1.23
2020-05-10T11:37:21.267552: Step 369, loss_total = 4.30, acc = 80.80%, sec/batch = 1.23
2020-05-10T11:37:28.863091: Step 374, loss_total = 4.20, acc = 82.14%, sec/batch = 1.25
2020-05-10T11:37:36.594195: Step 379, loss_total = 2.96, acc = 87.05%, sec/batch = 1.23
2020-05-10T11:37:43.859151: Step 384, loss_total = 4.24, acc = 78.57%, sec/batch = 1.23
2020-05-10T11:37:51.127640: Step 389, loss_total = 4.08, acc = 79.46%, sec/batch = 1.25
2020-05-10T11:37:59.199318: Step 394, loss_total = 3.76, acc = 83.48%, sec/batch = 1.25
2020-05-10T11:38:07.150536: Step 399, loss_total = 3.65, acc = 83.04%, sec/batch = 1.24
2020-05-10T11:38:15.290589: Step 404, loss_total = 4.06, acc = 82.14%, sec/batch = 1.23
2020-05-10T11:38:22.652271: Step 409, loss_total = 3.69, acc = 82.59%, sec/batch = 1.25
2020-05-10T11:38:30.074432: Step 414, loss_total = 3.74, acc = 82.14%, sec/batch = 1.23
2020-05-10T11:38:38.004456: Step 419, loss_total = 3.02, acc = 87.95%, sec/batch = 1.25
2020-05-10T11:38:46.111891: Step 424, loss_total = 2.99, acc = 84.82%, sec/batch = 1.23
2020-05-10T11:38:53.875687: Step 429, loss_total = 3.51, acc = 84.38%, sec/batch = 1.25
2020-05-10T11:39:01.295813: Step 434, loss_total = 3.15, acc = 84.38%, sec/batch = 1.25
2020-05-10T11:39:08.637353: Step 439, loss_total = 2.75, acc = 88.39%, sec/batch = 1.23
2020-05-10T11:39:16.412170: Step 444, loss_total = 2.95, acc = 87.05%, sec/batch = 1.25
2020-05-10T11:39:24.904576: Step 449, loss_total = 3.59, acc = 88.39%, sec/batch = 1.25
2020-05-10T11:39:32.996414: Step 454, loss_total = 2.76, acc = 86.16%, sec/batch = 1.23
2020-05-10T11:39:40.680170: Step 459, loss_total = 3.05, acc = 86.61%, sec/batch = 1.23
2020-05-10T11:39:47.992985: Step 464, loss_total = 2.61, acc = 89.29%, sec/batch = 1.23
2020-05-10T11:39:55.862590: Step 469, loss_total = 2.21, acc = 87.05%, sec/batch = 1.25
2020-05-10T11:40:03.810228: Step 474, loss_total = 2.69, acc = 88.84%, sec/batch = 1.24
2020-05-10T11:40:11.169477: Step 479, loss_total = 2.50, acc = 88.39%, sec/batch = 1.23
2020-05-10T11:40:18.948448: Step 484, loss_total = 2.07, acc = 89.29%, sec/batch = 1.25
2020-05-10T11:40:27.056505: Step 489, loss_total = 1.74, acc = 91.52%, sec/batch = 1.25
2020-05-10T11:40:34.554737: Step 494, loss_total = 1.85, acc = 92.41%, sec/batch = 1.23
2020-05-10T11:40:42.329134: Step 499, loss_total = 1.13, acc = 94.64%, sec/batch = 1.25
2020-05-10T11:41:00.006928: Step 504, loss_total = 1.83, acc = 91.96%, sec/batch = 1.23
2020-05-10T11:41:07.333364: Step 509, loss_total = 1.92, acc = 92.41%, sec/batch = 1.23
2020-05-10T11:41:14.936333: Step 514, loss_total = 2.14, acc = 89.73%, sec/batch = 1.23
2020-05-10T11:41:22.627070: Step 519, loss_total = 1.64, acc = 92.41%, sec/batch = 1.23
2020-05-10T11:41:29.908684: Step 524, loss_total = 1.60, acc = 93.30%, sec/batch = 1.25
2020-05-10T11:41:37.348937: Step 529, loss_total = 1.39, acc = 94.64%, sec/batch = 1.25
2020-05-10T11:41:45.450826: Step 534, loss_total = 1.60, acc = 92.86%, sec/batch = 1.23
2020-05-10T11:41:53.277075: Step 539, loss_total = 1.35, acc = 94.20%, sec/batch = 1.23
2020-05-10T11:42:01.025282: Step 544, loss_total = 1.72, acc = 92.41%, sec/batch = 1.23
2020-05-10T11:42:08.319825: Step 549, loss_total = 0.94, acc = 95.54%, sec/batch = 1.23
2020-05-10T11:42:15.767286: Step 554, loss_total = 2.05, acc = 89.73%, sec/batch = 1.25
2020-05-10T11:42:24.047741: Step 559, loss_total = 1.92, acc = 88.84%, sec/batch = 1.25
2020-05-10T11:42:32.061478: Step 564, loss_total = 1.06, acc = 95.09%, sec/batch = 1.23
2020-05-10T11:42:39.903409: Step 569, loss_total = 1.10, acc = 95.54%, sec/batch = 1.23
2020-05-10T11:42:47.292294: Step 574, loss_total = 1.34, acc = 94.20%, sec/batch = 1.24
2020-05-10T11:42:55.122575: Step 579, loss_total = 1.51, acc = 92.41%, sec/batch = 1.25
2020-05-10T11:43:02.541123: Step 584, loss_total = 0.90, acc = 97.32%, sec/batch = 1.23
2020-05-10T11:43:10.364357: Step 589, loss_total = 1.07, acc = 94.20%, sec/batch = 1.24
2020-05-10T11:43:18.152342: Step 594, loss_total = 1.49, acc = 92.86%, sec/batch = 1.23
2020-05-10T11:43:25.431847: Step 599, loss_total = 1.40, acc = 94.64%, sec/batch = 1.23
2020-05-10T11:43:33.509187: Step 604, loss_total = 2.06, acc = 92.41%, sec/batch = 1.25
2020-05-10T11:43:41.740506: Step 609, loss_total = 1.49, acc = 90.18%, sec/batch = 1.25
2020-05-10T11:43:49.348058: Step 614, loss_total = 1.18, acc = 93.75%, sec/batch = 1.23
2020-05-10T11:43:56.658836: Step 619, loss_total = 1.43, acc = 93.30%, sec/batch = 1.23
2020-05-10T11:44:04.398964: Step 624, loss_total = 0.73, acc = 97.77%, sec/batch = 1.25
2020-05-10T11:44:12.782010: Step 629, loss_total = 1.33, acc = 95.09%, sec/batch = 1.25
2020-05-10T11:44:20.670744: Step 634, loss_total = 0.86, acc = 95.09%, sec/batch = 1.25
2020-05-10T11:44:28.168978: Step 639, loss_total = 1.46, acc = 92.86%, sec/batch = 1.23
2020-05-10T11:44:35.924196: Step 644, loss_total = 1.17, acc = 93.75%, sec/batch = 1.25
2020-05-10T11:44:44.171648: Step 649, loss_total = 0.93, acc = 96.43%, sec/batch = 1.23
2020-05-10T11:44:51.623032: Step 654, loss_total = 0.94, acc = 95.98%, sec/batch = 1.25
2020-05-10T11:44:59.503751: Step 659, loss_total = 1.62, acc = 92.41%, sec/batch = 1.25
2020-05-10T11:45:08.514174: Step 664, loss_total = 0.86, acc = 96.43%, sec/batch = 1.23
2020-05-10T11:45:16.777853: Step 669, loss_total = 0.99, acc = 95.54%, sec/batch = 1.25
2020-05-10T11:45:24.650997: Step 674, loss_total = 1.19, acc = 95.09%, sec/batch = 1.23
2020-05-10T11:45:32.055500: Step 679, loss_total = 0.71, acc = 96.88%, sec/batch = 1.23
2020-05-10T11:45:39.649446: Step 684, loss_total = 0.79, acc = 95.98%, sec/batch = 1.25
2020-05-10T11:45:48.066795: Step 689, loss_total = 0.98, acc = 96.43%, sec/batch = 1.23
2020-05-10T11:45:55.924290: Step 694, loss_total = 0.92, acc = 96.88%, sec/batch = 1.23
2020-05-10T11:46:03.344405: Step 699, loss_total = 0.66, acc = 97.77%, sec/batch = 1.23
2020-05-10T11:46:11.055222: Step 704, loss_total = 0.70, acc = 97.77%, sec/batch = 1.23
2020-05-10T11:46:18.565500: Step 709, loss_total = 0.65, acc = 96.43%, sec/batch = 1.23
2020-05-10T11:46:26.404796: Step 714, loss_total = 1.16, acc = 93.75%, sec/batch = 1.25
2020-05-10T11:46:34.059208: Step 719, loss_total = 1.05, acc = 93.75%, sec/batch = 1.25
2020-05-10T11:46:41.574609: Step 724, loss_total = 0.93, acc = 95.54%, sec/batch = 1.23
2020-05-10T11:46:50.088750: Step 729, loss_total = 0.62, acc = 96.88%, sec/batch = 1.25
2020-05-10T11:46:58.352460: Step 734, loss_total = 1.52, acc = 95.98%, sec/batch = 1.25
2020-05-10T11:47:06.303679: Step 739, loss_total = 0.60, acc = 98.66%, sec/batch = 1.25
2020-05-10T11:47:13.989397: Step 744, loss_total = 0.62, acc = 97.32%, sec/batch = 1.25
2020-05-10T11:47:21.380321: Step 749, loss_total = 0.86, acc = 95.54%, sec/batch = 1.23
2020-05-10T11:47:29.674730: Step 754, loss_total = 0.47, acc = 98.66%, sec/batch = 1.25
2020-05-10T11:47:37.969619: Step 759, loss_total = 0.68, acc = 98.21%, sec/batch = 1.25
2020-05-10T11:47:45.936524: Step 764, loss_total = 0.85, acc = 96.88%, sec/batch = 1.23
2020-05-10T11:47:53.497249: Step 769, loss_total = 0.48, acc = 98.21%, sec/batch = 1.23
2020-05-10T11:48:00.776776: Step 774, loss_total = 0.53, acc = 97.77%, sec/batch = 1.24
2020-05-10T11:48:09.072674: Step 779, loss_total = 0.98, acc = 94.64%, sec/batch = 1.24
2020-05-10T11:48:17.391832: Step 784, loss_total = 0.85, acc = 96.43%, sec/batch = 1.25
2020-05-10T11:48:25.343082: Step 789, loss_total = 0.40, acc = 97.77%, sec/batch = 1.25
2020-05-10T11:48:32.981905: Step 794, loss_total = 0.56, acc = 97.77%, sec/batch = 1.23
2020-05-10T11:48:40.308310: Step 799, loss_total = 0.70, acc = 98.21%, sec/batch = 1.23
Training over. It costs 20.70 minutes

  

4、项目源代码

  需要加微信:wuzaipei

以上是关于python cnn tensorflow 车牌识别 网络模型的主要内容,如果未能解决你的问题,请参考以下文章

用于车牌识别的预训练 LeNet 模型

multi-digit街景识别代码到中文车牌识别的迁移

Tensorflow CNN入门

Python+Keras+TensorFlow车牌识别

Tensorflow 上带有回归输出的 CNN 图像识别

建立车牌识别模型的最佳方法 (ALPR)