与CPTN（文字识别网络）作斗争的记录

CTPN是什么

CTPN结合CNN与LSTM深度网络，CTPN是从Faster R-CNN改进而来，能有效的检测出复杂场景的横向分布的文字，效果如图1，是目前比较好的文字检测算法。详细解释：传送门

image.png
说人话：文字识别的前期工作，需要把图片中的文字区域定位出来，然后可以做适当的裁切作进一步的文字识别工作！

本次要实现的项目

地址如下（别着急克隆）：

https://github.com/eragonruan/text-detection-ctpn

image.png

项目要实现的效果

image.png
image.png

第一步--克隆项目

我克隆的是下面这个项目，他有训练好的权重文件：

https://github.com/GuoYi0/text_detection

第二步--开始跑demo

结果碰到了下面这个错误：

ImportError: cannot import name 'bbox'

回去找issue，果然有人跟我一样，才看半截就开跑
issue地址：https://github.com/eragonruan/text-detection-ctpn/issues/59
下面有人给出了答案：

image.png

照着做，再跑一次，但是，，，还是

image.png

another error：

Invalid argument: ValueError: Buffer dtype mismatch, expected 'int_t' but got 'long long'

再去找解决方案。找到了下面这个：
issue地址：https://github.com/eragonruan/text-detection-ctpn/issues/59
解决方案如下（我只是部分修改，原作者是仅使用CPU执行）

thanks to the author and [#43](https://github.com/eragonruan/text-detection-ctpn/issues/43) zhao181

my environment is:
windows10 ,
python3.6 ,
tensorflow1.3 ,
vs2015(ps:vs2013 not support python3.6 when compile)

step 1:make some change
change "np.int_t " to "np.intp_t" in line 25 of the file lib\utils\cython_nms.pyx
otherwise appear " ValueError: Buffer dtype mismatch, expected 'int_t' but got 'long long' " in step 6.

step 2:updata c file
execute:cd your_dir\text-detection-ctpn-master\lib\utils
execute:cython bbox.pyx
execute:cython cython_nms.pyx

step 3:builf setup file as setup_new.py
import numpy as np
from distutils.core import setup
from Cython.Build import cythonize
from distutils.extension import Extension
numpy_include = np.get_include()
setup(ext_modules=cythonize("bbox.pyx"),include_dirs=[numpy_include])
setup(ext_modules=cythonize("cython_nms.pyx"),include_dirs=[numpy_include])

step 4:build .pyd file
execute:python setup_new.py install
copy bbox.cp36-win_amd64.pyd and cython_nms.cp36-win_amd64.pyd to your_dir\text-detection-ctpn-master\lib\utils

step 5:make some change
(1) Set "USE_GPU_NMS " in the file \ctpn\text.yml as "False"
(2) Set the "_*C.USE_GPU_NMS" in the file \lib\fast_rcnn\config.py as "False";
(3) Comment out the line "from lib.utils.gpu_nms import gpu_nms" in the file \lib\fast_rcnn\nms_wrapper.py;
(4) Comment out the line "from . import gpu_nms" in the file \lib\utils_*init**.py;
(5) change "base_name = image_name.split('/')[-1]" to "base_name = image_name.split('\')[-1]" in line 24 of the file ctpn\demo.py

step 6:run demo
execute:cd your_dir\text-detection-ctpn-master
execute:python ./ctpn/demo.py

关键改动部分

step 1:make some change
change "np.int_t " to "np.intp_t" in line 25 of the file lib\utils\cython_nms.pyx
otherwise appear " ValueError: Buffer dtype mismatch, expected 'int_t' but got 'long long' " in step 6.

我再跑一次测试！
结果，，，

image.png

错误信息如下：

Loading network VGGnet_test...  Restoring from checkpoints/VGGnet_fast_rcnn_iter_50000.ckpt... done
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Demo for E:\business\recognition\text_detection\text-detection-ctpn\data\demo\010.png
Traceback (most recent call last):
  File "./ctpn/demo.py", line 101, in <module>
    ctpn(sess, net, im_name)
  File "./ctpn/demo.py", line 61, in ctpn
    draw_boxes(img, image_name, boxes, scale)
  File "./ctpn/demo.py", line 26, in draw_boxes
    with open('data/results/' + 'res_{}.txt'.format(base_name.split('.')[0]), 'w') as f:
OSError: [Errno 22] Invalid argument: 'data/results/res_E:\\business\\recognition\\text_detection\\text-detection-ctpn\\data\\demo\\010.txt'

定位这个问题花了我一点时间，各种猜想，最后再回头详细读一下这个错误信息，发现了问题所在，最后一句！参数错误！
回去改源码：

 base_name = image_name.split('\\')[-1]
    with open('data\\results\\' + 'res_{}.txt'.format(base_name.split('.')[0]), 'w') as f:

只想说一句，Windows的“\”真是。。。

image.png
我再跑！功夫不负有心人，输出结果如下
image.png
查看输出结果目录
image.png
接下来就是去研究原理和源码了，这个得放一段落，先把业务完成先。。。
image.png

再补一句，对这个感兴趣的还可以去研究下下面这个项目，提供了数据集等

https://github.com/QAlexBall/Faster_RCNN_for_TextDetection