TY - GEN
T1 - An enhanced MobileNetV4 architecture with CBAM for accurate classification mushroom toxicity
AU - Marasin, Alexandrio Kharisma Putra
AU - Utaminingrum, Fitri
AU - Fauzi, Mochammad Ali
N1 - Publisher Copyright:
© 2026 SPIE.
PY - 2026/4/15
Y1 - 2026/4/15
N2 - The morphological resemblance between edible and toxic mushroom species presents significant public health challenges, contributing to thousands of annual poisoning incidents globally through visual classification errors. This study proposes an enhanced MobileNetV4 architecture integrating Convolutional Block Attention Module (CBAM) with optimized reduction ratio for accurate mushroom toxicity classification on mobile devices. Using a dataset of 2,000 mushroom images across two classes (edible and poisonous), we systematically evaluated six model variants combining MobileNetV4 Small/Large with CBAM at different reduction ratios. Results demonstrate that MobileNetV4 Large+CBAM 16 achieves 96.90% accuracy with only 13.27M parameters, while maintaining real-time inference speed of 29.7 ms per image. The integration of CBAM with reduction ratio 16 effectively enhances feature representation for distinguishing morphologically similar species, outperforming baseline MobileNetV4 and existing lightweight architectures. This work enables practical deployment of accurate mushroom toxicity classification systems on resource-constrained devices for food safety applications.
AB - The morphological resemblance between edible and toxic mushroom species presents significant public health challenges, contributing to thousands of annual poisoning incidents globally through visual classification errors. This study proposes an enhanced MobileNetV4 architecture integrating Convolutional Block Attention Module (CBAM) with optimized reduction ratio for accurate mushroom toxicity classification on mobile devices. Using a dataset of 2,000 mushroom images across two classes (edible and poisonous), we systematically evaluated six model variants combining MobileNetV4 Small/Large with CBAM at different reduction ratios. Results demonstrate that MobileNetV4 Large+CBAM 16 achieves 96.90% accuracy with only 13.27M parameters, while maintaining real-time inference speed of 29.7 ms per image. The integration of CBAM with reduction ratio 16 effectively enhances feature representation for distinguishing morphologically similar species, outperforming baseline MobileNetV4 and existing lightweight architectures. This work enables practical deployment of accurate mushroom toxicity classification systems on resource-constrained devices for food safety applications.
KW - Convolutional Block Attention Module (CBAM)
KW - Deep Learning
KW - Food Safety
KW - MobileNetV4
KW - Mushroom Toxicity Classification
UR - https://www.scopus.com/pages/publications/105039500498
U2 - 10.1117/12.3111427
DO - 10.1117/12.3111427
M3 - Conference contribution
AN - SCOPUS:105039500498
T3 - Proceedings of SPIE - The International Society for Optical Engineering
BT - International Conference on Pattern Recognition and Image Analysis, PRIA 2025
A2 - Ma, Jixin
A2 - Fournier-Viger, Philippe
A2 - Zheng, Qian
A2 - Jain, Deepak Kumar
PB - SPIE
T2 - 2025 International Conference on Pattern Recognition and Image Analysis, PRIA 2025
Y2 - 26 December 2025 through 28 December 2025
ER -