Long-distance pipeline weld detection system and method based on CNN (convolutional neural network)

Abstract

The invention discloses a long-distance pipeline weld detection system and method based on a CNN (convolutional neural network). The system comprises a C1 layer feature map module, a S2 layer featuremap module and a C3 layer feature map module, wherein the C1 layer feature map module is used for convolving an input image with three trainable filters, and generating three feature maps on the C1 layer after convolving; the S2 layer feature map module is used for weighting and biasing the feature maps of the C1 layer, and then three feature maps of the S2 layers are obtained through a sigmoid number; the C3 layer feature map module is used for filtering the maps of the S2 layer feature map module to obtain feature maps of the C3 layer. The CNN is used, so that accuracy rate of weld judgmentis increased; equipment with the CNN is used for detecting long-distance pipeline welds, and accordingly, risks of manual detection are avoided.

Description

technical field [0001] The invention relates to a pipeline weld seam detection system, and more particularly to a long-distance pipeline weld seam detection system based on a convolutional neural network. Background technique [0002] In recent years, with the massive data generated by the rapid development of the Internet and the rapid development of computer hardware (CPU, GPU, etc.) Outstanding achievements in natural language processing, speech recognition and other fields. As an important part of deep learning, convolutional neural network (CNN) has been widely used in image processing due to its unique structural advantages. Convolutional neural networks (CNN) consist of one or more convolutional layers and a fully connected layer at the top, and include related weights and pooling layers. This structure enables CNN to utilize input data two-dimensional structure. Compared with other deep architectures, convolutional neural networks have shown excellent results in i...

AI Technical Summary

Problems solved by technology

[0004] At present, manual inspection is mostly used for long-distance pipeline welding seam inspection, and there are three major disadvantages: first, the pipeline is very narrow in some places, and the operators cannot reach it; second, the workload is large, and manual inspection takes a long time and is expensive; Space operations, frequent occurrence of hypoxia, dizziness, fatigue and other conditions for operators, there is a very high risk of life

Images