Minghui Wu

Minghui Wu

Auto-generated profile, based on publicly available information

Minghui Wu

School of mechanical and automotive engineering, Shanghai University of Engineering Science, Shanghai 201620, China

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 May 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
XRXinyue Ren
Xinyue Ren
WZWeiwei Zhang
Weiwei Zhang
MWMinghui Wu
Minghui Wu
KYKaixin Yang
Kaixin Yang
XWXiaolan Wang
Xiaolan Wang
Considering the low coverage of roadside cooperative devices at the current time, automated driving should detect all road markings relevant to driving safety, such as traffic signs that tend to be of great variety but are fewer in number. In this work, we propose an innovative few-shot object detection framework, namely Meta-YOLO, whose challenge is to generalize to the unseen classes by using only a few seen classes. Simply integrating the YOLO mechanism into a meta-learning pipeline will encounter problems in terms of computational efficiency and mistake detection. Therefore, we construct a two-stage meta-learner ℱ model that can learn the learner initialization, the learner update direction and learning rate, all in a single meta-learning process. To facilitate deep networks with learning, the fidelity features of the targets improve the performance of meta-learner ℱ, but we also design a feature decorrelation module (FDM), which firstly transforms non-linear features into computable linear features based on RFF, and secondly perceives and removes global correlations by iteratively saving and reloading the features and sample weights of the model. We introduce a three-head module to learn global, local and patch correlations with the category detection result outputted by the aggregation in meta-learner ℱ, which endows a multi-scale ability with detector ϕ. In our experiments, the proposed algorithm outperforms the three benchmark algorithms and improves the mAP of few-shot detection by 39.8%.

19 November 2018 in MATEC Web of Conferences
Verified
Journal Article
Open access
PSPeidi Shao
Peidi Shao
MWMinghui Wu
Minghui Wu
XWXianwei Wang
Xianwei Wang
JZJun Zhou
Jun Zhou
SLSheng Liu
Sheng Liu
The identification and extraction of tea buds is the key technology for the development of automated tea picking robots. Machine vision technology is an effective tool for tea bud recognition. In this paper, the tea tree leaves in the tea garden picking period are taken as research objects, and the research experiments are carried out from the aspects of tea image collection, image enhancement, image segmentation, edge detection, binarization and foreground extraction. After continuous exploration and research, the HSI color model is finally selected. After the S factor was used to grayscale the tea image, the improved K-means algorithm was used to identify and separate the tea shoots. The experimental results show that the improved K-means algorithm has a good effect on the segmentation of young leaves in tea images. This study can provide reference and reference for tea bud recognition algorithm.