Yolo v3-based Asian elephant identification method and system

Abstract

The invention discloses a Yolo v3-based Asian elephant identification method and system, and relates to the technical field of image identification. The method comprises the steps of constructing a Yolo v3-based Asian elephant identification model for training, verification and testing: collecting field Asian elephant image data as a data set, and dividing data set categories through WordTree, wherein the data set categories comprise the first-layer category, the second-layer category and the third-layer category, the first-layer category is an Asian elephant, the second-layer category is an RGB Asian elephant and an infrared Asian elephant, and the third-layer category is a full target category, a head category and a hip category of the RGB Asian elephant and a full target_i category, a head_i category and a hip_i category of the infrared Asian elephant; marking each Asian elephant image according to the divided categories through WordTree, dividing the marked data set into a training set, a verification set and a test set, and training, verifying and testing the Asian elephant identification model through the training set, the verification set and the test set; and carrying out Asian elephant identification by using the Asian elephant identification model.

Description

technical field [0001] The invention discloses a method and a system, and relates to the technical field of image recognition, in particular to a yolov3-based Asian elephant recognition method and system. Background technique [0002] Due to population increase, economic development and other reasons, land use patterns have undergone tremendous changes. A large number of natural forests outside the reserve (partially resting forests) and land suitable for forests have been reclaimed and planted into economic forests such as rubber, tea, and coffee, causing Asian elephants to The suitable habitat has been greatly reduced, coupled with the increase in the population of Asian elephants, the total amount of food they need has increased. In order to meet the needs of food, some Asian elephants have left the forest and moved to farmland. The range of activities has expanded year by year, and human-elephant conflicts continue. upgrade. However, the existing monitoring and early wa...

AI Technical Summary

Problems solved by technology

However, the existing monitoring and early warning of Asian elephants mostly rely on manual identification technology. Through manual selection and identification of front-end equipment data, it is a conventional method to determine the location, trajectory and regularity of Asian elephants. However, due to the limitations of large data volume, manpower and time, Difficult to complete all-round identification work in a short time

Images