Tensorflow-based CNN+LSTM trained with CTC-loss for OCR

Overview

Overview

This collection demonstrates how to construct and train a deep, bidirectional stacked LSTM using CNN features as input with CTC loss to perform robust word recognition.

The model is a straightforward adaptation of Shi et al.'s CRNN architecture (arXiv:1507.0571). The provided code downloads and trains using Jaderberg et al.'s synthetic data (IJCV 2016), MJSynth.

Notably, the model achieves a lower test word error rate (1.82%) than CRNN when trained and tested on case-insensitive, closed vocabulary MJSynth data.

Written for Python 2.7. Requires TensorFlow >=1.10 (deprecation warnings exist for TF>1.10, but the code still works).

The model and subsequent experiments are more fully described in Weinman et al. (ICDAR 2019)

Structure

The model as built is a hybrid of Shi et al.'s CRNN architecture (arXiv:1507.0571) and the VGG deep convnet, which reduces the number of parameters by stacking pairs of small 3x3 kernels. In addition, the pooling is also limited in the horizontal direction to preserve resolution for character recognition. There must be at least one horizontal element per character.

Assuming one starts with a 32x32 image, the dimensions at each level of filtering are as follows:

Layer Op KrnSz Stride(v,h) OutDim H W PadOpt
1 Conv 3 1 64 30 30 valid
2 Conv 3 1 64 30 30 same
Pool 2 2 64 15 15
3 Conv 3 1 128 15 15 same
4 Conv 3 1 128 15 15 same
Pool 2 2,1 128 7 14
5 Conv 3 1 256 7 14 same
6 Conv 3 1 256 7 14 same
Pool 2 2,1 256 3 13
7 Conv 3 1 512 3 13 same
8 Conv 3 1 512 3 13 same
Pool 3 3,1 512 1 13
9 LSTM 512
10 LSTM 512

To accelerate training, a batch normalization layer is included before each pooling layer and ReLU non-linearities are used throughout. Other model details should be easily identifiable in the code.

The default training mechanism uses the ADAM optimizer with learning rate decay.

Differences from CRNN

Deeper early convolutions

The original CRNN uses a single 3x3 convolution in the first two conv/pool stages, while this network uses a paired sequence of 3x3 kernels. This change increases the theoretical receptive field of early stages of the network.

As a tradeoff, we omit the computationally expensive 2x2x512 final convolutional layer of CRNN. In its place, this network vertically max pools over the remaining three rows of features to collapse to a single 512-dimensional feature vector at each horizontal location.

The combination of these changes preserves the theoretical receptive field size of the final CNN layer, but reduces the number of convolution parameters to be learned by 15%.

Padding

Another important difference is the lack of zero-padding in the first convolutional layer, which can cause spurious strong filter responses around the border. By trimming the first convolution to valid regions, this model erodes the outermost pixel of values from the response filter maps (reducing height from 32 to 30 and reducing the width by two pixels).

This approach seems preferable to requiring the network to learn to ignore strong Conv1 responses near the image edge (presumably by weakening the power of filters in subsequent convolutional layers).

Batch normalization

We include batch normalization after each pair of convolutions (i.e., after layers 2, 4, 6, and 8 as numbered above). The CRNN does not include batch normalization after its first two convolutional stages. Our model therefore requires greater computation with an eye toward decreasing the number of training iterations required to reach converegence.

Subsampling/stride

The first two pooling stages of CRNN downsample the feature maps with a stride of two in both spatial dimensions. This model instead preserves sequence length by downsampling horizontally only after the first pooling stage.

Because the output feature map must have at least one timeslice per character predicted, overzealous downsampling can make it impossible to represent/predict sequences of very compact or narrow characters. Reducing the horizontal downsampling allows this model to recognize words in narrow fonts.

This increase in horizontal resolution does mean the LSTMs must capture more information. Hence this model uses 512 hidden units, rather than the 256 used by the CRNN. We found this larger number to be necessary for good performance.

Training

To completely train the model, you will need to download the mjsynth dataset and pack it into sharded TensorFlow records. Then you can start the training process, a tensorboard monitor, and an ongoing evaluation thread. The individual commands are packaged in the accompanying Makefile.

make mjsynth-download
make mjsynth-tfrecord
make train &
make monitor &
make test

To monitor training, point your web browser to the url (e.g., (http://127.0.1.1:8008)) given by the Tensorboard output.

Note that it may take 4-12 hours to download the complete mjsynth data set. A very small set (0.1%) of packaged example data is included; to run the small demo, skip the first two lines involving mjsynth.

With a GeForce GTX 1080, the demo takes about 20 minutes for the validation character error to reach 45% (using the default parameters); at one hour (roughly 7000 iterations), the validation error is just over 20%.

With the full training data, by one million iterations the model typically converges to around 5% training character error and 27.5% word error.

Checkpoints

Pre-trained model checkpoints at DOI:11084/23328 are used to produce results in the following paper:

Weinman, J. et al. (2019) Deep Neural Networks for Text Detection and Recognition in Historical Maps. In Proc. ICDAR.

Testing

The evaluate script (src/evaluate.py) streams statistics for one batch of validation (or evaluation) data. It prints the iteration, evaluation batch loss, label error (percentage of characters predicted incorrectly), and the sequence error (percentage of words—entire sequences—predicted incorrectly).

The test script (src/test.py) tallies statistics, finally normalizing for all data. It prints the loss, label error, total number of labels, sequence error, total number of sequences, and the label error rate and sequence error rate.

Validation

To see the output of a small set of instances, the validation script (src/validation.py) allows you to load a model and read an image one at a time via the process's standard input and print the decoded output for each. For example

cd src ; python validate.py < ~/paths_to_images.txt

Alternatively, you can run the program interactively by typing image paths in the terminal (one per line, type Control-D when you want the model to run the input entered so far).

Configuration

There are many command-line options to configure training parameters. Run train.py or test.py with the --help flag to see them or inspect the scripts. Model parameters are not command-line configurable and need to be edited in the code (see src/model.py).

Dynamic training data

Dynamic data can be used for training or testing by setting the --nostatic_data flag.

You can use the --ipc_synth boolean flag [default=True] to determine whether to use single-threaded or a buffered, multiprocess synthesis.

The --synth_config_file flag must be given with --nostatic_data.

The MapTextSynthesizer library supports training with dynamically synthesized data. The relevant code can be found within MapTextSynthesizer/tensorflow/generator

Using a lexicon

By default, recognition occurs in "open vocabulary" mode. That is, the system observes no constraints on producing the resulting output strings. However, it also has a "closed vocabulary" mode that can efficiently limit output to a given word list as well as a "mixed vocabulary" mode that can produce either a vocabulary word from a given word list (lexicon) or a non-vocabulary word, depending on the value of a prior bias for lexicon words.

Using the closed or mixed vocabulary modes requires additional software. This repository is connected with a fork of Harald Scheidl's CTCWordBeamSearch, obtainable as follows:

git clone https://github.com/weinman/CTCWordBeamSearch
cd CTCWordBeamSearch
git checkout var_seq_len

Then follow the build instructions, which may be as simple as running

cd cpp/proj
./buildTF.sh

To use, make sure CTCWordBeamSearch/cpp/proj (the directory containing TFWordBeamSearch.so) is in the LD_LIBRARY_PATH when running test.py or validate.py (in this repository).

API Notes

This version uses the TensorFlow (v1.14) Dataset for fast I/O. Training, testing, validation, and prediction use a custom Estimator.

Citing this work

Please cite the following paper if you use this code in your own research work:

@inproceedings{ weinman19deep,
    author = {Jerod Weinman and Ziwen Chen and Ben Gafford and Nathan Gifford and Abyaya Lamsal and Liam Niehus-Staab},
    title = {Deep Neural Networks for Text Detection and Recognition in Historical Maps},
    booktitle = {Proc. IAPR International Conference on Document Analysis and Recognition},
    month = {Sep.},
    year = {2019},
    location = {Sydney, Australia},
    doi = {10.1109/ICDAR.2019.00149}
} 

Acknowledgment

This work was supported in part by the National Science Foundation under grant Grant Number 1526350.

Owner
Jerod Weinman
Associate Professor of Computer Science
Jerod Weinman
Demo processor to illustrate OCR-D Python API

ocrd_vandalize/ Demo processor to illustrate the OCR-D/core Python API Description :TODO: write docs :) Installation From PyPI pip3 install ocrd_vanda

Konstantin Baierer 5 May 05, 2022
OpenGait is a flexible and extensible gait recognition project

A flexible and extensible framework for gait recognition. You can focus on designing your own models and comparing with state-of-the-arts easily with the help of OpenGait.

Shiqi Yu 335 Dec 22, 2022

Installations for running keras-theano on GPU Upgrade pip and install opencv2 cd ~ pip install --upgrade pip pip install opencv-python Upgrade keras

Berat Kurar Barakat 14 Sep 30, 2022
learn how to use Gesture Control to change the volume of a computer

Volume-Control-using-gesture In this project we are going to learn how to use Gesture Control to change the volume of a computer. We first look into h

Diwas Pandey 49 Sep 22, 2022
Thresholding-and-masking-using-OpenCV - Image Thresholding is used for image segmentation

Image Thresholding is used for image segmentation. From a grayscale image, thresholding can be used to create binary images. In thresholding we pick a threshold T.

Grace Ugochi Nneji 3 Feb 15, 2022
A curated list of papers and resources for scene text detection and recognition

Awesome Scene Text A curated list of papers and resources for scene text detection and recognition The year when a paper was first published, includin

Jan Zdenek 43 Mar 15, 2022
Repository of conference publications and source code for first-/ second-authored papers published at NeurIPS, ICML, and ICLR.

Repository of conference publications and source code for first-/ second-authored papers published at NeurIPS, ICML, and ICLR.

Daniel Jarrett 26 Jun 17, 2021
[BMVC'21] Official PyTorch Implementation of Grounded Situation Recognition with Transformers

Grounded Situation Recognition with Transformers Paper | Model Checkpoint This is the official PyTorch implementation of Grounded Situation Recognitio

Junhyeong Cho 18 Jul 19, 2022
This project proposes a camera vision based cursor control system, using hand moment captured from a webcam through a landmarks of hand by using Mideapipe module

This project proposes a camera vision based cursor control system, using hand moment captured from a webcam through a landmarks of hand by using Mideapipe module

Chandru 2 Feb 20, 2022
Use Convolutional Recurrent Neural Network to recognize the Handwritten line text image without pre segmentation into words or characters. Use CTC loss Function to train.

Handwritten Line Text Recognition using Deep Learning with Tensorflow Description Use Convolutional Recurrent Neural Network to recognize the Handwrit

sushant097 224 Jan 07, 2023
aardio的opencv库

opencv_aardio dll库下载地址:https://github.com/xuncv/opencv-plugin/releases import cv2 img = cv2.imread("./images/Lena.jpg",1) img = cv2.medianBlur(img,5)

71 Dec 31, 2022
一键翻译各类图片内文字

一键翻译各类图片内文字 针对群内、各个图站上大量不太可能会有人去翻译的图片设计,让我这种日语小白能够勉强看懂图片 主要支持日语,不过也能识别汉语和小写英文 支持简单的涂白和嵌字

574 Dec 28, 2022
MONAI Label is a server-client system that facilitates interactive medical image annotation by using AI.

MONAI Label is a server-client system that facilitates interactive medical image annotation by using AI. It is an open-source and easy-to-install ecosystem that can run locally on a machine with one

Project MONAI 344 Dec 23, 2022
Some codes from PyImageSearch course's and external projects.

👨‍💻 Some codes and projects 👨‍💻 💡 Technologies 📜 Projects 📍 Chrome Dinosaur Controller 📦 Script 📍 Coins Counter 📦 Script 🤓 Author Lucas Biv

Lucas Bivar 25 Oct 24, 2021
EQFace: An implementation of EQFace: A Simple Explicit Quality Network for Face Recognition

EQFace: A Simple Explicit Quality Network for Face Recognition The first face recognition network that generates explicit face quality online.

DeepCam Shenzhen 141 Dec 31, 2022
PianoVisuals - Create background videos synced with piano music using opencv

Steps Record piano video Use Neural Network to do body segmentation (video matti

Solbiati Alessandro 4 Jan 24, 2022
✌️Using this you can control your PC/Laptop volume by Hand Gestures created with Python.

Hand Gesture Volume Controller ✋ Hand recognition 👆 Finger recognition 🔊 you can decrease and increase volume Demo Code Firstly I have created a Mod

Abbas Ataei 19 Nov 17, 2022
Provides OCR (Optical Character Recognition) services through web applications

OCR4all As suggested by the name one of the main goals of OCR4all is to allow basically any given user to independently perform OCR on a wide variety

174 Dec 31, 2022
Text layer for bio-image annotation.

napari-text-layer Napari text layer for bio-image annotation. Installation You can install using pip: pip install napari-text-layer Keybindings and m

6 Sep 29, 2022
Train custom VR face tracking parameters

Pal Buddy Guy: The anipal's best friend This is a small script to improve upon the tracking capabilities of the Vive Pro Eye and facial tracker. You c

7 Dec 12, 2021