lyl8213/Plate_Recognition-LPRnet
Tensorflow based. Use CNN and CTC loss. It is a light network for Plate Recognition
Python
Issues
- 0
可提供50w+中文车牌数据
#33 opened by 123Vincent2018 - 0
车牌检测和车牌识别商用级别数据集
#32 opened by daijie1223 - 2
- 2
UnicodeEncodeError
#5 opened by mochechan - 0
麻烦请问转pb模型的输出节点名称是什么呢?
#31 opened by GongYuxiu - 0
可提供数据支持
#30 opened by daijie1223 - 0
Character segmenation
#29 opened - 4
How to test single lpr plate .
#8 opened by pankajbadatia - 0
how to train/test
#28 opened by Hedibouchelliga - 0
Error after train
#27 opened by pendex900x - 0
Full dataset
#26 opened by mrmohammadi - 0
识别时,相同词很大概率会当成合并为一个如何解决
#25 opened by lightning0016 - 0
有没有论文里面的数据集,我想训练一下
#24 opened by liuzhuang1024 - 8
为什么训练到30epoch左右就出现train_cost为0,而且val_cost没有减少?而且使用model8.24best里的模型进行测试时都是出现长度不一致,得不到测试结果?
#14 opened by YDS-FOREVER - 0
请问CTC对数概率log_prob可以转换为车牌的识别率么?
#23 opened by huchenw - 0
- 3
len(original_list) 50 len(detected_list) 1 test and detect length desn't match
#16 opened by zhangwang0537 - 0
transfer learning
#20 opened by sinadunk23 - 0
- 0
Prediction script
#17 opened by poltavski - 2
Is possible to train two line number plate?
#15 opened by bluesy7585 - 1
Train on different type of licence plate?
#13 opened by jihadbourassi - 2
if len(original_list) != len(detected_list):
#12 opened by xinyuegtxy - 6
你训练的model acc有多少?
#9 opened by lanybass - 6
how to train
#1 opened by panlanlan - 5
can the label_len be any length ?
#4 opened by tuqingyun - 1
- 3
The width and Height is misplaced
#10 opened by MisayaZ - 5
Question: Output layer for graph freeze?
#6 opened by lmmaia - 0
- 1
the recognition speed
#2 opened by HouBiaoLiu