BriefGPT.xyz
Mar, 2022
自标准化密度地图（SNDM）用于微生物物体计数
Self-Normalized Density Map (SNDM) for Counting Microbiological Objects
HTML
PDF
Krzysztof M. Graczyk, Jarosław Pawłowski, Sylwia Majchrowska, Tomasz Golan
TL;DR
研究了基于密度图方法的微生物物体计数的统计特性，并利用深度神经网络应用了bootstrap和MC dropout两种统计学方法，提出了一种自归一化模块来纠正模型预测的缺陷并提高精度。该模型称为自归一化密度图模型，有效性与Faster和Cascade R-CNN检测器相当。
Abstract
The statistical properties of the
density map
(DM) approach to counting
microbiological objects
on images are studied in detail. The DM is given by U$^2$-Net. Two
→