Dynamic vision sensor background activity noise filtering method and system based on hash function and medium

• Main Authors: WANG LEI, ZHANG JIANFENG, LI SHIMING, DING DONG, GUO SHASHA, WANG YONGWEN, KANG ZIYANG, WANG SHUQUAN, ZHANG LIMENG, XU WEIXIA
• Format: Patent
• Language: Chinese; English
• Published: 17.04.2020
• Subjects: ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PICTORIAL COMMUNICATION, e.g. TELEVISION

The invention discloses a dynamic vision sensor background activity noise filtering method and system based on a hash function and a medium. The method comprises the steps: an event event output by aDVS is received; the event event is compressed from the high-dimensional vector by adopting a hash function to obtain a real value; information coding is carried out on the real value and the real value is mapped to a target position in an m list; whether the original value of the target position is greater than or equal to a preset threshold Thr or not is judged, if so, that the event event is judged to be a noise event and is filtered, or otherwise, the event event is judged to be a real event; and the value mapped to the target position in the m list is updated. According to the method, thebackground activity noise of the dynamic vision sensor can be filtered, memory resources occupied by the method are fewer than those occupied by a traditional nearest neighbor filter, meanwhile, theprocessing speed is ensured