Region-based Convolutional Neural Networks for Text Label Detection in Patent Images
Written by : Taehoon Kim, jamesk1228@gmail.com, Sogang University, Seoul, Republic of Korea
Objective: Detect digits from black & white patent images.
Prerequsites: python3, tensorflow, scikit-image, numpy, opencv-python (Recommand using anaconda)
How-to:
- Clone this git repository
- For pre-trained MNIST Handwritten digit Convolutional Neural Network model, go to https://drive.google.com/open?id=0B7TzarHysU5fNXlwMF9Cb0RTV1U
- Unzip pre-trained model and place it under root git repository.
- Place your custom image in "images/"
- To run the code in Jupyter Notebook, use main.ipynb instead of main.py.
- You can find our raw value and coordinates of labels in patent image from "labels/image_name_label.txt"
Corresponding Paper: on process
References:
https://github.com/FraPochetti/ImageTextRecognition
[1] Christoph Riedl, Richard Zanibbi, Marti A. Hearst, Siyu Zhu, Michael Menietti, Jason Crusan, Ivan Metelsky, Karim R. Lakhani, Detecting figures and part labels in patents: competition-based development of graphics recognition algorithms International Journal on Document Analysis and Recognition (IJDAR), Volume 19, Issue 2, pp. 155–172, 2016
[2] Ross Girshick, Jeff Donahue, Trevor Darrell, Region-Based Convolutional Networks for Accurate Object Detection and Segmentation, IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume: 38, Issue 1, pp. 142–158, 2015
[3] Max Jaderberg, Karen Simonyan, Andrea Vedaldi, Andrew Zisserman, Reading Text in the Wild with Convolutional Neural Networks, International Journal of Computer Vision, Volume 116, Issue 1, pp. 1-20, 2016
[4] Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich, Going Deeper with Convolutions, The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1-9, 2015
[5] Xiang Zhang, Junbo Zhao, Yann LeCun, Character-level Convolutional Networks for Text Classification, Advances in Neural Information Processing Systems 28 (NIPS 2015)
[6] Nobuyuki Otsu, "A threshold selection method from gray-level histograms". IEEE Trans. Sys., Man., Cyber. 9 (1), pp. 62–66, 1979