Building construction vehicle identification method based on deep learning

Abstract

The invention discloses a building construction vehicle identification method based on deep learning, and the method comprises the steps of 1, collecting the data of each construction vehicle type used on a construction site, and carrying out the classification; 2, according to the collected data of each type of construction vehicles, establishing a construction vehicle database of a correspondingtype; 3, converting a data set in each type of construction vehicle database, building a corresponding multi-label convolutional neural network model, and carrying out parameter initialization on themodel; 4, training a multi-label convolutional neural network model; 5, judging whether the trained multi-label convolutional neural network model meets the requirements or not; 6, testing and evaluating the trained multi-label convolutional neural network model; 7, judging whether the test evaluation meets the standard or not; and 8, outputting a final multi-label convolutional neural network model and parameters, wherein the model and the parameters are applied to the follow-up identification of the construction vehicles. The method is high in identification accuracy and is more convenientfor the construction vehicle management on the construction site.

Description

technical field [0001] The invention belongs to the technical field of building construction, in particular to a method for identifying building construction vehicles based on deep learning. Background technique [0002] The construction industry is an industry that revolves around the design, construction, decoration and management of buildings. Urban architecture is an important part of the city, and architecture is not just an artificial work for people to stay, rest, and entertain. It is related to our economy, culture, and life in a large way. Therefore, the construction industry is particularly important. [0003] Many special construction vehicles are used in the construction process, such as cranes, forklifts, trucks, concrete mixer trucks, material transport vehicles and many other types of construction vehicles. These construction vehicles are different from ordinary household vehicles. The existing models cannot be directly used to identify construction vehicles...

AI Technical Summary

Problems solved by technology

These construction vehicles are different from ordinary household vehicles. The existing models cannot be directly used to identify construction vehicles, and they cannot be replaced to manage various types of construction vehicles on the construction site.