Learning to count small and clustered objects with application to bacterial colonies

2026-04-21

AI Summary

Automated bacterial colony counting faces significant challenges, including small object sizes, high-density clustering, high annotation costs, and limited cross-species generalization. To address these issues, this work proposes ACFamNet and its enhanced variant, ACFamNet Pro, which introduce an aligned region-of-interest (ROI) pooling mechanism integrated with refined feature engineering, multi-head attention, and residual connections. These innovations substantially improve counting accuracy for small targets and enhance cross-species generalization. Evaluated within a density map regression framework using five-fold cross-validation, ACFamNet Pro achieves an average normalized absolute error of 9.64%, representing improvements of 2.23% and 12.71% over ACFamNet and FamNet, respectively.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Automated bacterial colony counting from images is an important technique to obtain data required for the development of vaccines and antibiotics. However, bacterial colonies present unique machine vision challenges that affect counting, including (1) small physical size, (2) object clustering, (3) high data annotation cost, and (4) limited cross-species generalisation. While FamNet is an established object counting technique effective for clustered objects and costly data annotation, its effectiveness for small colony sizes and cross-species generalisation remains unknown. To address the first three challenges, we propose ACFamNet, an extension of FamNet that handles small and clustered objects using a novel region of interest pooling with alignment and optimised feature engineering. To address all four challenges above, we introduce ACFamNet Pro, which augments ACFamNet with multi-head attention and residual connections, enabling dynamic weighting of objects and improved gradient flow. Experiments show that ACFamNet Pro achieves a mean normalised absolute error (MNAE) of 9.64% under 5-fold cross-validation, outperforming ACFamNet and FamNet by 2.23% and 12.71%, respectively.

Problem

• bacterial colony counting
• small object counting
• object clustering
• cross-species generalisation
• data annotation cost

Innovation

• region of interest pooling with alignment
• multi-head attention
• residual connections
• small object counting
• cross-species generalisation