Luggage re-identification model training method and luggage re-identification method

Abstract

The invention discloses a luggage re-identification model training and luggage re-identification method, and the method comprises the following steps: S1, obtaining video monitoring data, and extracting a luggage image of the video monitoring data through a target detection model; S2, labeling the luggage image to obtain a luggage image data set; S3, dividing the luggage image data set into a training set and a verification set; S4, training a luggage re-identification model by using the training set, evaluating the luggage re-identification model by using the verification set, and screening out the luggage re-identification model with the optimal prediction performance; S5, obtaining similarity distribution between the luggage images by using the luggage re-identification model obtained in the step S4, and determining a discrimination threshold according to the similarity distribution; and S6, inputting any two luggage images into the luggage re-identification model to predict whetherthe two luggage images are from the same luggage or not. According to the invention, the defects of high cost and low security of a traditional luggage identification tracking scheme are overcome, the portability is strong, and the identification accuracy is high.

Description

technical field [0001] The present invention relates to the technical field of object re-identification in computer vision, and more specifically, to a luggage re-identification model training and a luggage re-identification method. Background technique [0002] Existing luggage tracking scheme generally uses RFID (radio frequency identification) technology to track luggage, and the scheme includes the following steps: 1. The RFID (radio frequency identification) card reader in each area reads the RFID (radio frequency identification) ) tag and send it to the data summarization unit, and the RFID (radio frequency identification) tag is attached to the luggage; 2. The data summarization unit in each location sums up the luggage information in each area and sends it to the data collection unit. Processing system; 3. The data processing system generates whereabouts information according to all the luggage information received, and sends the whereabouts information to the inform...

AI Technical Summary

Problems solved by technology

[0007] The present invention provides a luggage weight recognition model training and a luggage weight recognition method in order to overcome the defects of high cost, low security, and troublesome implementation of luggage tracking and identification in the prior art

Images