Point element annotation configuration method

• Main Authors: XU JIAQI, WU XIAOFENG, REN XIAOLU, ZHANG YONG, LIU CUI
• Format: Patent
• Language: Chinese; English
• Published: 22.12.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention relates to a point element annotation configuration method, and belongs to the discrete combination optimization problem. Aiming at the problem that a standard artificial bee colony algorithm cannot adapt to a solving process of point element annotation configuration, the method comprises the following steps of: firstly, coding based on a selected annotation candidate model, and randomly initializing a bee colony to adapt to a discretization solving process; secondly, learning excellent individuals by adopting a learning-based updating strategy in an employed bee stage so as to improve the diversity of solutions; in an observation bee stage, a more desirable solution is selected for updating based on a selection operation of annotation similarity, and updating is realized by adopting a dynamic neighborhood search method, so that a solution space is fully explored, and the search capability of an artificial bee colony algorithm is improved; and finally, a Metropolis acceptance criterion is used f