Title:
IMAGE-BASED TRAFFIC SIGNAL LIGHT DETECTION METHOD AND APPARATUS, ELECTRONIC DEVICE AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/139302
Kind Code:
A1
Abstract:
The invention relates to the technical field of artificial intelligence, specifically to image detection/image classification/neural network technology, and relates in particular to an image-based traffic information light detection method and apparatus, an electronic device, and a computer readable storage medium. The invention applies to areas such as assisted driving and navigation, automatic driving, and smart traffic, thus promoting smart city construction. The image-based traffic information light detection method comprises: acquiring an original image; reducing the resolution of the original image to obtain a rough detection image; performing target region detection on the rough detection image by using a rough detection convolutional neural network so as to obtain region information of a target region; matching the region information of the target region with the original image and obtaining from the original image an original image block corresponding to the target region; and using a fine detection convolutional neural network to carry out traffic signal light region detection on the original image block so as to acquire position and category information in respect of the traffic signal light region.
Inventors:
ZHONG ZHIQUAN (CN)
ZHUANG BOJIN (CN)
WANG SHAOJUN (CN)
ZHUANG BOJIN (CN)
WANG SHAOJUN (CN)
Application Number:
PCT/CN2020/121870
Publication Date:
July 15, 2021
Filing Date:
October 19, 2020
Export Citation:
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G06N3/04; G06K9/00; G06N3/08
Foreign References:
| CN111160206A | 2020-05-15 | |||
| CN108875608A | 2018-11-23 | |||
| US20190385005A1 | 2019-12-19 |
Attorney, Agent or Firm:
SHENZHEN CHINA INNOVATION SOUTH INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: