Research on convolutional neural network for video classification

Zhou, Yaru

doi:10.1117/12.2624727

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An improved version of LeNet, LeNet-5 emerged in 1998. It can extract translation-invariant features of images and achieve excellent outcomes in handwritten digit recognition, which significantly boosted the development of convolutional neural networks in the classification of two-dimensional images [14]. So far, CNN has achieved breakthrough results in image recognition and classification [15,16].…”

Section: Fault Diagnosis Process Of Cnnmentioning

confidence: 99%

Applications of convolutional neural networks in fault diagnosis of mechanical-electrical-hydraulic systems: a review

Liu,

Zhou,

Yue

et al. 2024

Fourth International Conference on Mechanical Engineering, Intelligent Manufacturing, and Automation Technology (MEMAT 2023)

View full text Add to dashboard Cite

The mechanical-electrical-hydraulic systems are developing towards automation and intelligence, which raise higher requirements for accurate and smart fault diagnosis of the core component in these systems. Traditional diagnostic methods not only rely too much on a priori knowledge for feature extraction but also have limited computational ability for massive data, which can't satisfy the demand for intelligent fault diagnosis. Thanks to the deep network structure, deep learning (DL) can automatically extract the deep implicit feature information from multiple data without a priori knowledge and has an immense advantage in realizing complicated fault diagnosis. As a typical representative of deep learning, Convolutional neural network (CNN) can directly process images and has made significant achievements in image recognition and classification, therefore scholars have conducted extensive research on CNN for fault diagnosis. Based on the above-mentioned content, this review sorts out the current research works employing the 1DCNN model, 2DCNN model and hybrid model of CNN for fault diagnosis in mechanical-electrical-hydraulic systems, and then summarizes the advantages and disadvantages of the three CNN models, respectively. Finally, we discuss the future development directions and challenges of fault diagnosis with CNN.

show abstract

Section: Fault Diagnosis Process Of Cnnmentioning

confidence: 99%