large-field ddPCR_YOLOv5

Published: 17 December 2024| Version 1 | DOI: 10.17632/f6rjrn2w7g.1
Contributor:
星宇

Description

The dataset and code used in this study are crucial for advancing the accurate detection of positive microchambers in large-field ddPCR imaging. The provided dataset includes annotated ddPCR images in YOLO format. The code includes the improved YOLOv5 model, which integrates BiFPN, GhostConv, C3Ghost modules, SimAM attention mechanism, and network pruning, among other custom modifications. Additionally, scripts for model training and evaluation, as well as the results of all comparative experiments, are provided. All the code and datasets also include an operation interface developed using PyQt5, which facilitates image processing and result analysis for users.

Files

Institutions

  • University of Shanghai for Science and Technology

Categories

Image Processing, Polymerase Chain Reaction, Automatic Target Recognition, Deep Learning

Licence