公开(公告)号：EP3869398A3

公开(公告)日：2022-01-12

申请号：EP21180877.9

申请日：2021-06-22

申请人： BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO. LTD.

发明人： ZHANG, Chengquan ,  EN, Mengyi ,  HUANG, Ju ,  XIE, Qunyi ,  QIN, Xiameng ,  YAO, Kun ,  HAN, Junyu ,  LIU, Jingtuo ,  DING, Errui

IPC分类号： G06K9/00 ,  G06K9/62 ,  G06K9/32 ,  G06K9/46

摘要： A method and apparatus for processing an image, a device and a storage medium. An implementation of the method includes: acquiring a template image, the template image including at least one region of interest; determining a first feature map corresponding to each region of interest in the template image; acquiring a target image; determining a second feature map of the target image; and determining at least one region of interest in the target image according to the first feature map and the second feature map.

公开(公告)号：EP3923185A2

公开(公告)日：2021-12-15

申请号：EP21202754.4

申请日：2021-10-14

申请人： Beijing Baidu Netcom Science and Technology Co., Ltd.

发明人： YU, Yuechen ,  ZHANG, Chengquan ,  LI, Yulin ,  ZHANG, Xiaoqiang ,  HUANG,, Ju ,  QIN, Xiameng ,  YAO, Kun ,  LIU, Jingtuo ,  HAN, Junyu ,  DING, Errui

IPC分类号： G06K9/00 ,  G06K9/62

摘要： Provided are an image classification method and apparatus, an electronic device and a storage medium, relating to the field of artificial intelligence and, in particular, to computer vision and deep learning. The method includes inputting (S101, S201, S301) a to-be-classified document image into a pretrained neural network and obtaining a feature submap of each text box of the to-be-classified document image by use of the neural network; inputting (S102, S202, S302) the feature submap of each text box, a semantic feature corresponding to preobtained text information of each text box and a position feature corresponding to preobtained position information of each text box into a pretrained multimodal feature fusion model and fusing, by use of the multimodal feature fusion model, the three into a multimodal feature corresponding to each text box; and classifying (S103) the to-be-classified document image based on the multimodal feature corresponding to each text box. The semantic feature and position feature in the document image are well used so that the object of improving the classification accuracy of the document image is achieved.

公开(公告)号：EP3836016A1

公开(公告)日：2021-06-16

申请号：EP21162692.4

申请日：2021-03-15

申请人： Beijing Baidu Netcom Science and Technology Co., Ltd.

发明人： LI, Yulin ,  HUANG, Ju ,  QIN, Xiameng ,  HAN, Junyu

IPC分类号： G06K9/00 ,  G06K9/36 ,  G06K9/46 ,  G06K9/62

摘要： The present disclosure discloses a method, apparatus, device and storage medium for recognizing a bill image and a computer program product, relates to the fields of artificial intelligence deep learning and image processing. A specific implementation is: performing text detection on a bill image, and determining an attribute information set and a relationship information set of each text box of at least two text boxes in the bill image; determining a type of the text box and an associated text box that has a structural relationship with the text box based on the attribute information set and the relationship information set of the text box; and extracting structured bill data of the bill image, based on the type of the text box and the associated text box that has the structural relationship with the text box. The solution of embodiments of the present disclosure can support automatic recognition of a variety of bill images in different formats, and the recognition process does not require use of a template, which improves the versatility and accuracy of bill image recognition.