Bridge disease real-time detection method based on model pruning

Abstract

The invention discloses a bridge disease real-time detection method based on model pruning, and the method comprises the following steps: firstly obtaining a bridge apparent damage image through an unmanned plane, and dividing a data set into a training set, a verification set and a test set according to a proportion; increasing the number of data sets by applying an image enhancement method; constructing a yolov3 network framework, adding L1 regularization constraints of gamma parameter items to each batch of normalization layers (BN) of the network, setting a sparsity adjustment strategy ofprogressive local attenuation, and training the reconstructed network to obtain a training model; after the gamma parameters of all the trained BN layers are obtained, performing model pruning of thelayers and the channels according to the sizes of the gamma parameters between the layers and in the layers, deleting the channel with the small weight, and obtaining a pruned model; and carrying outbridge apparent disease automatic identification by using the trained model. The invention is high in efficiency and low in cost, and compared with a traditional manual label marking training method,the method has more obvious automation and high efficiency.

Description

technical field [0001] The invention relates to an interactive technology of civil engineering and artificial intelligence, in particular to a real-time detection method for bridge diseases based on model pruning. Background technique [0002] During the service of the bridge, the bridge pier and the bridge deck are prone to cracks under long-term loads. As one of the main diseases of the concrete structure, the crack will affect the durability and bearing capacity of the structure; in addition, the anchorage section of the stay cable, etc. It is also prone to rust and corrosion, which may affect the bearing capacity of the stay cables, thereby affecting the service safety of the bridge. Regular inspection of the bridge structure can detect cracks, cable anchorage modification and other diseases in time, prevent and maintain them as early as possible, and improve the service life of the bridge structure. The traditional method of bridge structure apparent disease detection ...

AI Technical Summary

Problems solved by technology

Compared with the traditional detection algorithm, these methods have greatly improved the detection accuracy, but there is a big gap between the detection speed and the traditional algorithm, which is relatively slow.

Images