Self-Normalized Density Map (SNDM) for Counting Microbiological Objects

TL;DR

This paper introduces the Self-Normalized Density Map (SNDM), a novel neural network model that self-corrects its density map predictions for more accurate counting of microbiological objects, supported by statistical analysis.

Contribution

The paper proposes a self-normalization module integrated into the density map network, improving accuracy and consistency in object counting tasks compared to previous models.

Findings

SNDM outperforms the original density map model in object counting accuracy.

Both bootstrap and MC dropout methods yield consistent statistical results for SNDM.

SNDM's efficiency is comparable to detector-based models like Faster and Cascade R-CNN.

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 statistical methods for deep neural networks are utilized: the bootstrap and the Monte Carlo (MC) dropout. The detailed analysis of the uncertainties for the DM predictions leads to a deeper understanding of the DM model's deficiencies. Based on our investigation, we propose a self-normalization module in the network. The improved network model, called \textit{Self-Normalized Density Map} (SNDM), can correct its output density map by itself to accurately predict the total number of objects in the image. The SNDM architecture outperforms the original model. Moreover, both statistical frameworks -- bootstrap and MC dropout -- have consistent statistical results for SNDM, which were not observed in the original model. The SNDM efficiency is comparable with the detector-base models, such as Faster and Cascade R-CNN detectors.