Prof. Haixia Pan Profile

Prof. Haixia Pan

at Beihang Univ

SPIE Involvement:

Author

Publications (2)

This will count as one of your downloads.

You will have access to both the presentation and article (if available).

DOWNLOAD NOW

This content is available for download via your institution's subscription. To access this item, please sign in to your personal account.

Email or Username Forgot your username?

Password Forgot your password?

Show

Keep me signed in

No SPIE account? Create an account

Proceedings Article | 18 November 2024 Paper

A structured pruning method based on MobileNetV3

Haixia Pan, Shiheng Wang, Huolong Ye, Sinan Lin

Proceedings Volume 13403, 134030Z (2024) https://doi.org/10.1117/12.3052027

KEYWORDS: Tunable filters, Neural networks, Convolution, Education and training, Visual process modeling, Performance modeling, Structural design, Reflection, Computer vision technology, Batch normalization

Read Abstract +

In recent years, deep neural networks have made significant advancements, surpassing traditional models in computer vision. However, the vast number of parameters in deep models poses challenges such as increased network size, sluggish computation, high computational costs, and extensive storage requirements, thus hindering deployment on resource-constrained platforms like mobile devices. This paper focuses on compressing convolutional neural networks to address these challenges. This paper proposes a structured pruning method that integrates MobileNet features to prune the MobileNetV3-Large model, resulting in a more compact and efficient model. The approach involves sparse regularization training to obtain a sparser network model, which is followed up by structured pruning. This pruning is performed by leveraging the product of the sparsity values of convolutional layers and the scaling factors of batch normalization layers to identify and remove redundant filters. Experiments were conducted on the CIFAR-10 dataset. The experimental results demonstrate that the proposed compression method effectively reduces model parameters, decreases model size, minimizes runtime memory usage, and diminishes computational operations, while maintaining comparable performance levels despite the compression. Specifically, for MobileNetV3-Large, the model's parameter count is reduced by 40%, and computational operations are reduced by 35%, all while maintaining the same level of accuracy.

Proceedings Article | 10 July 2024 Paper

YOLO-DCANet: seismic velocity spectrum picking method based on deformable convolution and attention mechanism

Sibo Wang, Haixia Pan, Weifeng Geng, Ce Bian, Xiaosai Zhang, Biao Dong

Proceedings Volume 13223, 132232C (2024) https://doi.org/10.1117/12.3035559

KEYWORDS: Data modeling, Deformation, Object detection, Convolution, Feature extraction, Deep learning, Visualization, Machine learning

Read Abstract +

Seismic velocity spectrum picking is an important task in seismic exploration. With the development of geological exploration technology, the data of seismic velocity spectrum has experienced explosive growth. Manual picking of seismic velocity spectrum alone gradually cannot meet the needs of practical production. In order to improve efficiency and reduce time costs, there is an urgent need to find a fast and effective method to achieve an automated velocity spectrum picking process. This paper proposes an intelligent picking method based on neural networks. The method abstracts the task of seismic velocity spectrum picking into a visual task, processes it, and outputs candidate boxes through model training. The center points of the candidate boxes are used as the "time-velocity" sequences of seismic velocity spectrum images. To improve the accuracy of model training, a YOLO-based improved YOLO-DCANet network is proposed. In the backbone stage, deformable convolutional modules are introduced, and a Coordinate Attention (CA) mechanism is integrated at different stages of the process. Meanwhile, to learn the features of energy clusters at different levels in the image, a multi-scale model is introduced to match energy cluster information at different levels. The experiments show that the mAP accuracy of the proposed YOLO-DCANET is improved by about 8% compared with the traditional two-stage model, and it’s also improved by about 5% compared with the original YOLO series model. It shows that the seismic velocity spectrum picking algorithm proposed in this paper is significantly improved compared with the traditional deep learning method. In addition, the generalization test on other working areas proves that the proposed method has strong generalization ability and robustness.

My Library

You currently do not have any folders to save your paper to! Create a new folder below.

Folder Name

Folder Description

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks. You are receiving this notice because your organization may not have SPIE eBooks access.*

*Shibboleth/Open Athens users─please sign in to access your institution's subscriptions.

To obtain this item, you may purchase the complete book in print or electronic format on SPIE.org.

ORGANIZATIONAL
Sign in with credentials provided by your organization.

Organizational Username

Organizational Password

Show/Hide Password

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members:

Non-members: ADD TO CART

Keywords/Phrases

Search In:

Publication Years