W
Wenxuan Shi
Researcher at Wuhan University
Publications - 11
Citations - 208
Wenxuan Shi is an academic researcher from Wuhan University. The author has contributed to research in topics: Image quality & Feature (computer vision). The author has an hindex of 5, co-authored 11 publications receiving 175 citations.
Papers
More filters
Journal ArticleDOI
Real-Time Fabric Defect Detection Using Accelerated Small-Scale Over-Completed Dictionary of Sparse Coding:
TL;DR: A hardware accelerated algorithm based on a small-scale over-completed dictionary (SSOCD) via sparse coding (SC) method, which is realized on a parallel hardware platform (TMS320C6678) and shows that the proposed algorithm can run with high parallel efficiency and meets the real-time requirements of industrial inspection.
Journal ArticleDOI
No-reference image quality assessment based on hybrid model
TL;DR: A computational algorithm based on hybrid model to automatically extract vision perception features from raw image patches is proposed, which demonstrates very competitive quality prediction performance of the proposed method.
Journal ArticleDOI
Blind image quality assessment in multiple bandpass and redundancy domains
TL;DR: A novel method that employs both bandpass and redundancy domains to acquire the complementary features in multiple color spaces is proposed and is very competitive against other BIQA methods and has good generalization ability.
Journal ArticleDOI
Refining deep convolutional features for improving fine-grained image recognition
TL;DR: A fine-grained image recognition framework is proposed by exploiting CNN as the raw feature extractor along with several effective methods including a feature encoding method, a feature weighting method, and a strategy to better incorporate information from multi-scale images to further improve recognition ability.
Journal ArticleDOI
Classification of foreign fibers using deep learning and its implementation on embedded system
TL;DR: An embedded system based on field programmable gate array + digital signal processor to recognize and remove foreign fibers mixed in cotton is proposed and a convolution neural network mode is developed to validate the classification of the suspected targets from the detection subsystem, to improve the detection reliability.