scispace - formally typeset
Z

Zhongping Zhang

Researcher at Los Alamos National Laboratory

Publications -  22
Citations -  469

Zhongping Zhang is an academic researcher from Los Alamos National Laboratory. The author has contributed to research in topics: Inversion (meteorology) & Feature selection. The author has an hindex of 8, co-authored 21 publications receiving 259 citations. Previous affiliations of Zhongping Zhang include Boston University & University of Rochester.

Papers
More filters
Journal ArticleDOI

Automatic detection of particle size distribution by image analysis based on local adaptive canny edge detection and modified circular Hough transform.

TL;DR: An automatic image processing algorithm which is mainly based on local adaptive Canny edge detection and modified circular Hough transform is proposed which can utilize the local thresholds to detect particles from the images with different degrees of complexity.
Journal ArticleDOI

Data-Driven Seismic Waveform Inversion: A Study on the Robustness and Generalization

TL;DR: Zhang et al. as mentioned in this paper proposed a real-time data-driven technique called VelocityGAN, which is built on a generative adversarial network (GAN) and trained end-to-end to learn a mapping function from the raw seismic waveform data to the velocity image.
Posted Content

"Factual" or "Emotional": Stylized Image Captioning with Adaptive Learning and Attention

TL;DR: In this paper, a novel stylized image captioning model is proposed to generate stylized captions that have a specific style (e.g., humorous, romantic, positive, and negative) while describing the image content accurately.
Book ChapterDOI

“Factual” or “Emotional”: Stylized Image Captioning with Adaptive Learning and Attention

TL;DR: A novel stylized image captioning model that effectively takes factual and stylized knowledge into consideration and outperforms the state-of-the-art approaches, without using extra ground truth supervision is proposed.
Posted Content

Data-driven Seismic Waveform Inversion: A Study on the Robustness and Generalization

TL;DR: This article developed a real-time data-driven technique and it is called VelocityGAN, to reconstruct accurately the subsurface velocities and develops a transfer-learning strategy based on VelocityGAN to alleviate the generalization issue.