Patent
Complex image and text sequence identification method
Liu Shilin,He Hongjing,Chen Bingzhang,Wu Yunong,Yao Jia +4 more
- 15 Jun 2016
12
TL;DR: In this article, a sliding sampling box is used to perform sliding sampling on an image and text sequence to identify the sub images obtained through sampling by means of a CNN and outputting the characteristics to an RNN, wherein the RNN successively identifies the front part of each character, the back part of every character, numbers, letters, punctuation, or blank according to the input signal.
read more
Abstract: The invention relates to the image and text identification field, and specifically relates to a complex image and text sequence identification method. The complex image and text sequence identification method includes the steps: utilizing a sliding sampling box to perform sliding sampling on an image and text sequence to be identified; extracting the characteristics from the sub images obtained through sampling by means of a CNN and outputting the characteristics to an RNN, wherein the RNN successively identifies the front part of each character, the back part of each character, numbers, letters, punctuation, or blank according to the input signal; and successively recording and integrating the identification results for the RNN at each moment and acquiring the complete identification result, wherein the input signal for each moment for the RNN also includes the output signal of a recursion neural network for the last moment. The complex image and text sequence identification method can overcome the cutting problem of a complex image and text sequence, and can significantly improve the identification efficiency and accuracy for images and text.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Text recognition method and device
Yang Song,Wan Shaohua,Chen Zhijun +2 more
- 15 Feb 2017
TL;DR: In this paper, the authors proposed a text recognition method that combines CNN and RNN to achieve integral recognition of the text image to be recognized, so that the method avoids the problem that texts need to be separated character by character and then recognized character-by-character in the prior art which leads to accumulated errors, and the text recognition rate can be improved.
21
Patent
Image-based character recognition method
Xie Chang,Qian Haoran,Baohan Xu,Lu-Wang Tianyu +3 more
- 29 Jun 2018
TL;DR: In this paper, an image-based character recognition method is characterized by comprising steps as follows: an acquired image is divided into multiple areas; character areas in the multiple areas are detected to acquire the position information of the character areas and types of characters in the characters areas; the character regions are screened to determine the to-be-recognized character area.
14
Patent
Image identification method and apparatus
TL;DR: In this article, an image identification method consisting of extracting features of the to-be-identified image and inputting the features to a recurrent neural network to obtain one or more candidate regions corresponding to objects in the image and indicating the positions of the corresponding objects is presented.
11
Patent
Automatic invoice identification method based on deep learning
He Zhiyun,Luo Kun,Liang Xueying,Li Jing,Su Chao +4 more
- 15 Jun 2018
TL;DR: In this paper, an automatic invoice identification method based on deep learning is proposed, where images of invoices are collected; character cutting is carried out on the collected invoice images; character classification and grouping are carried out; and the grouped characters are imported into an AlexNet deep-learning network for training, and invoice identification is completed by the AlexNet network of which training is completed.
8
Patent
Neural network system based on reusing small-scale convolutional neural network module
Xing Wang,Jie Liang,Him Wai Ng +2 more
- 21 Nov 2017
TL;DR: In this article, a small-scale hardware convolutional neural network (CNN) was proposed for high-resolution input images, which is equivalent to the result of directly applying a large-scale CNN system to a high resolution input image.
7
References
Patent
System and method for modeling the neocortex and uses therefor
Douglas S. Greer
- 15 May 2007
TL;DR: In this article, a manifold association processor for a learning machine is presented, which uses a massive array of processing elements having local, recurrent connections to form global associations between functions defined on manifolds.
16
Patent
Mode identification method and device
Hu Xiaolin,Liang Ming +1 more
- 22 Jul 2015
TL;DR: In this article, the authors proposed a mode identification method based on a convolutional neural network with a recursion convolution layer, which is a neural network that total input is obtained with the combination of in-layer recursion input on the basis of feedforward input.
7
Patent
Image classification method based on locality preserving mapping and principal component analysis
Da Zhen,Wu Jiasong,Longyu Jiang,Youyong Kong,Yang Chunfeng,Shu Huazhong +5 more
- 02 Sep 2015
TL;DR: In this paper, an image classification method based on locality preserving mapping and principal component analysis (PCA) is proposed. And the method comprises the following steps: 1, inputting a training target; 2, establishing one layer or two layers of manifold learning networks, for each concrete network, learning and obtaining each layer filter; 3, establishing an output layer of the manifold learning network and acquiring a final characteristic vector; 4, inputing all the final characteristic vectors obtained from the step 3 into a support vector machine classifier so as to carry out training; 5, using a
2
Related Papers (5)
Chen Wang,Zhengzhen Liu,Xingyu Niu +2 more
- 09 May 2012
Chen Bingzhang,He Hongjing,Liu Shilin,Wu Yunong +3 more
- 20 Apr 2016