Title:
DETECTION METHOD AND APPARATUS BASED ON CONVOLUTIONAL NEURAL NETWORK, AND COMPUTER DEVICE AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2022/127043
Kind Code:
A1
Abstract:
A detection method and apparatus based on a convolutional neural network, and a computer device and a storage medium. The method belongs to the technical field of neural networks in artificial intelligence, and comprises: preprocessing an acquired target OCT image to obtain a plurality of feature maps of different scales; adjusting the plurality of feature maps of different scales according to a preset method so as to obtain first feature maps; calculating a response value of each pixel point in each first feature map by means of a preset algorithm, and marking the corresponding first feature map with a calculation result so as to obtain a second feature map; performing resolution restoration on each second feature map by means of an upsampling method, so as to obtain a target feature map; identifying a target region in the target feature map according to a region proposal network and the response value, so as to obtain an OCT image to be subjected to detection; and inputting the OCT image to be subjected to detection into a multilayer fully connected network so as to respectively determine the positions in which lesions are present, and the categories of the lesions. By means of the method, the efficiency and accuracy of detecting a lesion in an OCT image can be improved.
Inventors:
ZHANG CHENGFEN (CN)
LV BIN (CN)
LV CHUANFENG (CN)
LV BIN (CN)
LV CHUANFENG (CN)
Application Number:
PCT/CN2021/097190
Publication Date:
June 23, 2022
Filing Date:
May 31, 2021
Export Citation:
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G06T7/00
Foreign References:
| CN112541900A | 2021-03-23 | |||
| CN111667468A | 2020-09-15 | |||
| CN107369160A | 2017-11-21 | |||
| CN111615706A | 2020-09-01 | |||
| US20190325215A1 | 2019-10-24 |
Attorney, Agent or Firm:
SHENZHEN TALENT PATENT SERVICE (CN)
Download PDF: