|本期目录/Table of Contents|

[1]蒋冲宇,鲁统伟*,闵 峰,等.基于神经网络的发票文字检测与识别方法[J].武汉工程大学学报,2019,(06):586-590.[doi:10. 3969/j. issn. 1674-2869. 2019. 06. 013]
 JIANG Chongyu,LU Tongwei*,MIN Feng,et al.Invoice Text Detection and Recognition Based on Neural Network[J].Journal of Wuhan Institute of Technology,2019,(06):586-590.[doi:10. 3969/j. issn. 1674-2869. 2019. 06. 013]
点击复制

基于神经网络的发票文字检测与识别方法(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
期数:
2019年06期
页码:
586-590
栏目:
机电与信息工程
出版日期:
2021-01-24

文章信息/Info

Title:
Invoice Text Detection and Recognition Based on Neural Network
文章编号:
20190613
作者:
蒋冲宇鲁统伟*闵 峰熊寒颖胡记伟
武汉工程大学计算机科学与工程学院,湖北 武汉 430205
Author(s):
JIANG Chongyu LU Tongwei* MIN Feng XIONG HanyingHu Jiwei
School of Computer Science and Engineering, Wuhan Institute of Technology, Wuhan 430205, China
关键词:
印章发票颜色提取神经网络
Keywords:
seal invoicecolor extraction neural network
分类号:
TP391.4
DOI:
10. 3969/j. issn. 1674-2869. 2019. 06. 013
文献标志码:
A
摘要:
针对发票识别无法提取发票中被印章遮挡住的文本信息,提出了一种基于神经网络的发票文字检测与识别方法。首先使用轻量级深度神经网络检测出印章区域,利用区域的颜色信息分离印章和区域内文字信息,其中分离到的印章包含被印章覆盖的文字。然后对于印章,通过色彩阈值设定的方式提取被印章覆盖的字符。最后将区域内的文字和印章覆盖的文字合并,实现去除印章的目的。实验表明:算法有效减少计算量。算法在发票印章区域内文字检测准确率提高了53%、识别准确率提高了20%,实验证明了本文算法的可行和有效性。
Abstract:
Existing invoice identification methods have a deficiency in extracting text information covered by the seal. We propose a method for invoice text detection and recognition based on the neural network. Firstly, the seal’s area is detected by using a light-weight neural network. Then, the seal and the text are separated according to their color. Since the extracted seal’s area contains text information covered by the seal, a color threshold is used to extract the text under the seal. Finally, in order to remove the seal in the invoice, the text extracted in the seal’s area and the text covered by the seal are fused. Experimental results show that our method is computationally efficient. Additionally, the accuracy of text detection and recognition in the seal’s area are improved by 53% and 20%, respectively. It proves the feasibility and effectiveness of the proposed method.

参考文献/References:

[1] 江明,刘辉,黄欢. 图像二值化技术的研究[J]. 软件导刊, 2009(4):175-177. [2] 赵雪松,陈淑珍. 综合全局二值化与边缘检测的图像分割方法[J]. 计算机辅助设计与图形学学报,2001(2):118-121. [3] 刘仁军. 基于神经网络的室内场景的文字识别研究[D]. 武汉:武汉工程大学,2017. [4] 刘渝,张彦铎,鲁统伟. OCR中一种基于最小一乘法的连通域去噪方法[J]. 武汉工程大学学报, 2011,33(1):84-87. [5] 赵永涛,李志敏,王洪建,等. 印章识别中的图像预处理研究[J]. 仪器仪表学报,2004,25(S2):401-403. [6] 季婧婧,娄震. 基于二次分割的银行票据彩色印章的滤除[J]. 现代电子技术,2014,37(22):5-9. [7] LIU W,ANGUELOV D,ERHAN D, et al.Ssd:Single shot multibox detector[C]//European conference on computer vision.Cham:Springer,2016:21-37. [8] REDMON J,DIVVALA S,GIESHICK R,et al.You only look once:Unified,real-time object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.Las Vegas:IEEE,2016:779-788. [9] REN S,HE K M,GIRSHICK R,et al.Faster r-cnn:Towards real-time object detection with region proposal networks[C]//Proceedings of the IEEE conference on Computer Vision and Pattern Recongnition. Boston:IEEE,2015: 91-99. [10] TIAN Z,HUANG W L,HE T,et al.Detecting text in natural image with connectionist text proposal network[C]//European conference on computer vision.Cham:Springer,2016:56-72. [11] HOWARD A G,ZHU M,CHEN B,et al.Mobilenets:Efficient convolutional neural networks for mobile vision applications[C]//Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. Hawaii:IEEE,2017:1704-1709. [12] 欧阳欢. 多特征融合决策的发票印章识别[J]. 计算机工程与设计,2018,39(9):150-155. [13] 卢涛,章瑾,陈白帆,等. 多尺度自适应配准的视频超分辨率算法[J]. 武汉工程大学学报,2016,38(2):178-184. [14] 汪家明,卢涛. 基于多尺度残差深度神经网络的卫星图像超分辨率算法[J]. 武汉工程大学学报,2018,40(4):440-445. [15] SAINATH T N,KINGSBURY B,SAON G,et al.Deep convolutional neural networks for large-scale speech tasks[J]. Neural Networks,2015,64:39-48. [16] ERHAN D,SZRGRDY C,TOSHEV A,et al.Scalable object detection using deep neural networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.Columbus:IEEE,2014:2147-2154. [17] SZEGEDY C,REED S,ERHAN D,et al.Scalable,high-quality object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Columbus:IEEE,2014:1412-1441.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2019-05-21 基金项目:武汉工程大学研究生教育创新基金(CX2018202) 作者简介:蒋冲宇,硕士研究生。E-mail:[email protected] *通讯作者:鲁统伟,博士,副教授。E-mail:[email protected] 引文格式:蒋冲宇,鲁统伟,闵峰,等. 基于神经网络的发票文字检测与识别方法[J]. 武汉工程大学学报,2019,41(6):586-590.
更新日期/Last Update: 2020-01-20