Human body target detection method, device and computer readable storage medium

• Main Author: XIE CHANGYI
• Format: Patent
• Language: Chinese; English
• Published: 05.11.2019
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a human body target detection method. The method comprises the steps of obtaining a to-be-detected human body image; detecting the to-be-detected human body image by using theclassifier to obtain the position of a human body target area; wherein the process of obtaining the classifier comprises the following steps: S121, performing image training on a human body sample image set by utilizing a deep neural network algorithm to obtain the classifier; S122, carrying out human body target detection by utilizing a classifier; S123, judging whether the total amount of the detected human body images is greater than a preset threshold value or not; and S124, if the number of iterations is greater than the preset threshold value, determining a classifier obtained by performing image training by using a deep neural network algorithm on the basis of the classifier according to the detected human body image set as a new classifier, and entering S122 until the number of iterations reaches the first