Human leg detection method and device based on support vector description and readable medium

• Main Authors: YAO HANCHEN, PENG JIANWEI, WANG ZENGWEI, LIAN YANGLIN, XIA XUKE, LIN ZHIRONG, DAI HOUDE, ZHANG ZHEN, ZHU LIQI
• Format: Patent
• Language: Chinese; English
• Published: 12.08.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; PHYSICS

The invention discloses a human leg detection method and device based on support vector description and a readable medium, and relates to the field of human leg detection.The human leg detection method comprises the steps that two-dimensional laser point data, collected by a laser radar, of human legs are obtained, preliminary clustering is conducted on the two-dimensional laser point data through a k-nearest neighbor algorithm, and multiple candidate leg clusters are obtained; density features and geometric shape features of each candidate leg cluster are obtained through calculation based on the candidate leg clusters; calculating a leg support vector of each candidate leg cluster by adopting support vector description according to the density features and the geometrical shape features; and based on the leg support vectors of the candidate leg clusters, judging whether the two candidate leg clusters belong to the same pedestrian target cluster according to a decision function, and obtaining a clustering re