Short description

I am trying to train Tesseract on Akkadian language. The language-specific.sh script was modified accordingly. When converting the training text to TIFF images, the text2image program crashes.

Environment

• Tesseract Version: 3.04.01
• Commit Number: the standard package in Ubuntu, package version 3.04.01-4, commit unknown
• Platform: Linux ubuntu

1 In History, select multiple captures
2 Execute SHIFT+F10, Copy, FIlename
3 Inspect clipboard

Expected: multiple filenames
Observed: one filename - from focussed capture

Workaround: ??

Running paperless on python 3.8 (on ubuntu 20.04) with gunicorn 19.9.0 yields this error:

/usr/lib/python3.8/os.py:1023: RuntimeWarning: line buffering (buffering=1) isn't supported in binary mode, the default buffer size will be used
  return io.open(fd, *args, **kwargs)

This was resolved in gunicorn 20.0 per this issue and doing

From Issue #84, I did pod install, but facing following error:

ld: warning: directory not found for option '-F/Users/mahbub/Library/Developer/Xcode/DerivedData/NavigateMe-ehyamrhmeogwarbxhzbydagosybb/Build/Products/Debug-iphonesimulator/GPUImage'
ld: warning: directory not found for option '-F/Users/mahbub/Library/Developer/Xcode/DerivedData/NavigateMe-ehyamrhmeogwarbxhzbydagosybb/Build/P

我参考这个链接，yolov3-tiny.weights的这个模型用OpenCV的dnn是调用起来了的
https://github.com/spmallick/learnopencv/blob/master/ObjectDetection-YOLO/object_detection_yolo.cpp
但是直接替换模型路径为text.weights的输出不了文字区域坐标哦，反正前向传播是完成了的，net.forward(outs, getOutputsNames(net));就是没结果啊 不知道为啥。
还有请问下你的text.weights的直接输出不是文字外接矩吗？看你的代码的detector里面好像做了很多后处理啊，能稍微解释下整个流程吗？(yolov3的text.weights输出是什么，具体后面要做什么才能获得检测的文字的外接矩)

我看你的text_detect 有好多参

cvNet = cv.dnn.readNetFromTensorflow('ctpn.pb', 'protobuf.pbtxt')
cv2.error: OpenCV(4.1.1) C:\projects\opencv-python\opencv\modules\dnn\src\tensorflow\tf_io.cpp:54: error: (-2:Unspecified error) FAILED: ReadProtoFromTextFile(param_file, param). Failed to parse GraphDef file: protobuf.pbtxt in function 'cv::dnn::ReadTFNetParamsFromTextFileOrDie'

Does anyone have any idea how I can solve this pr

If I was to look at scanning several 100 pages what would be the best approach? In particular is there any benefit in scanning at higher dpi for better OCR accuracy but having lower resolution image stored? Can OCRmyPDF do this already itself or in combination with something?

Love the interface, looking for information on semi-basic questions.

1. This is a fairly fleshed out interface, are there any menu guides?
2. What/how is the default document storage location?
3. With the given setup/run sh files, how would I go about configuring a connection to my pre-existing database?

There's a problem with the formation of lines and paragraphs on this certain document.
It might be linked to the slanted nature of the text (slight rotation of the content on the input document).

Source File:

Output:
![image](https://user-images.githubusercontent.com/1119728/63240

Hi,
There is a typo in readme - demo section (should be _ rather than -)
is:
python3 run_demo_server.py --checkpoint-path /tmp/east_icdar2015_resnet_v1_50_rbox/
and should be:
python3 run_demo_server.py --checkpoint_path /tmp/east_icdar2015_resnet_v1_50_rbox/

also i had to change:
/tmp/east_icdar2015_resnet_v1_50_rbox/
into:
tmp/east_icdar2015_resnet_v1_50_rbox/

I think we should add a paper which is named "Detecting Text in Natural Scenes with
Stroke Width Transform",although,it is very early,lots of classical papers mention it.
address:https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/201020CVPR20TextDetection.pdf

建议精简版抛弃exe，直接打包成压缩文件。再带一个批处理，利用已经安装好的Java环境直接命令行运行。

I noticed that you used image height param format as the font size.
https://github.com/Belval/TextRecognitionDataGenerator/blob/33d8985521645280e102987e773bf1e424a045df/TextRecognitionDataGenerator/computer_text_generator.py#L14
In my test, image_font = ImageFont.truetype(font=font_size=500), no error was reported, but it was time consuming.
So I am confused, why set format, font_size