PSE Community.org

ISSN: 2818-4734
Volume: 5 (2026)
Table of Contents

LAPSE:2026.0415

Published Article

LAPSE:2026.0415

Hand-crafted Feature Fusion for Deep Learning-Based Instance Segmentation in Microfluidics

Wenle Xu, Lin Sheng, Qichen Shang, Mengqi Liu, Tong Qiu, Kai Wang, Guangsheng Luo

June 12, 2026

Abstract
High-throughput analysis of microfluidic droplets and bubbles is essential for chemical engineering but remains challenging due to the inherent loss of high-frequency details in standard deep learning models. This study proposes a novel Hand-crafted Feature Fusion framework that explicitly integrates physical priors, specifically Local Binary Patterns and Discrete Wavelet Transform, into a two-stage instance segmentation network. We design an adaptive attention-based fusion module embedded within both the Feature Pyramid Network and Region Proposal Network to synergize explicit texture cues with implicit semantic features. Validated on a large-scale dataset comprising over 64000 instances, our method achieves a test mAP of 0.808, significantly outperforming state-of-the-art architectures. Crucially, the framework effectively resolves the detection bottleneck for minute targets and elevates the small-object accuracy to 0.764, representing an improvement of nearly 20% over the baseline. This work demonstrates that incorporating physical priors offers a superior strategy for precise scientific image analysis compared to generic data-driven models.

Record ID

LAPSE:2026.0415

Keywords

Computer Vision, Hand-crafted Features, Instance Segmentation, Microfluidics

Subject

Modelling and Simulations

Suggested Citation

Xu W, Sheng L, Shang Q, Liu M, Qiu T, Wang K, Luo G. Hand-crafted Feature Fusion for Deep Learning-Based Instance Segmentation in Microfluidics. Systems and Control Transactions 5:1696-1704 (2026) https://doi.org/10.69997/sct.107493

Author Affiliations

Xu W: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China [ORCID]
Sheng L: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China
Shang Q: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China
Liu M: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China
Qiu T: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China
Wang K: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China
Luo G: Department of Chemical Engineering, Tsinghua University, Beijing, China. State Key Laboratory of Chemical Engineering and Low-carbon Technology, Tsinghua University, Beijing, China
[Login] to see author email addresses.

Journal Name

Systems and Control Transactions

Volume

First Page

1696

Last Page

1704

Year

2026

Publication Date

2026-06-12

DOI: