Experimental and Comparative Evaluation of an Integral-Based Squeeze Mechanism for Enhanced Feature Calibration in DenseNet121

Pandi Barita Nauli Simangunsong; Tuti  Andriani; Paska Marto  Hasugian

doi:10.52436/1.jutif.2026.7.4.5725

Authors

Pandi Barita Nauli Simangunsong Data Science, Universitas Katolik Santo Thomas, Indonesia
Tuti Andriani STMIK Mulia Darma, Indonesia
Paska Marto Hasugian Data Science, Universitas Katolik Santo Thomas, Indonesia

DOI:

https://doi.org/10.52436/1.jutif.2026.7.4.5725

Keywords:

Channel Attention Mechanism, Deep Learning, Feature Recalibration, Integral Based Attention, Squeeze-and-Excitation (SE) Block

Abstract

The Squeeze-and-Excitation (SE) Block, a concrete form of channel attention mechanisms, serves to flexibly readjust feature channels, with squeeze using Global Average Pooling for global context extraction across all channels. The limitation of the SE Block lies in discrete spatial merging, which causes important information to be lost. This research proposes an Integral-based SE Block by formulating the Squeeze operation as a continuous spatial integral to produce clearer channel feature merging. The IntSE block is implemented on the densenet121 architecture by replacing the discrete spatial SE without changing the structure of the backbone. The effectiveness of the architecture proposed in this study uses secondary datasets, namely Plant Village and HM10000, which were used for experiments during evaluation. The PlantVillage dataset consists of 12,246 tomato leaf data and consists of 10 classes, while the HAM10000 dataset consists of 10,015 data and has 7 classes. The results show that DenseNet121 with IntSE achieves an accuracy of 99.25% and an F1-score of 99.25% for the PlantVillage dataset, as well as an accuracy of 83.57% and an F1-score of 83.19% for the HAM10000 dataset. This experiment shows that densenet121 with IntSE outperforms basic densenet121, Discrete Spatial SE, and Attention CBAM in terms of accuracy and F1-Score. The ablation analysis shows that the formula of spatial squeeze integral performance is improved compared to global pooling and adaptive pooling. Statistical testing shows significant changes to state that the performance of the actual architecture.

Downloads

Download data is not yet available.

References

S. Khan, M. Sajjad, T. Hussain, A. Ullah, and A. S. Imran, “A review on traditional machine learning and deep learning models for WBCs classification in blood smear images,” IEEE Access, vol. 9, pp. 10657–10673, 2021, doi: 10.1109/ACCESS.2020.3048172.

T. Turay and T. Vladimirova, “Toward Performing Image Classification and Object Detection with Convolutional Neural Networks in Autonomous Driving Systems: A Survey,” IEEE Access, vol. 10, pp. 14076–14119, 2022, doi: 10.1109/ACCESS.2022.3147495.

M. M. Taye, “Theoretical Understanding of Convolutional Neural Network: Concepts, Architectures, Applications, Future Directions,” Computation, vol. 11, no. 3, 2023, doi: 10.3390/computation11030052.

S. Cong and Y. Zhou, A review of convolutional neural network architectures and their optimizations, vol. 56, no. 3. Springer Netherlands, 2023.

A. Gomaa and O. M. Saad, “Residual Channel-attention (RCA) network for remote sensing image scene classification,” Multimed. Tools Appl., vol. 84, no. 28, pp. 33837–33861, 2025, doi: 10.1007/s11042-024-20546-8.

M. B. Stuart, M. Davies, M. J. Hobbs, T. D. Pering, A. J. S. McGonigle, and J. R. Willmott, “High‐Resolution Hyperspectral Imaging Using Low‐Cost Components: Application within Environmental Monitoring Scenarios,” Sensors, vol. 22, no. 12, 2022, doi: 10.3390/s22124652.

S. G. S, T. Beucler, F. I. Tam, J. Runge, and A. Gerhardus, “Selecting robust features for machine-learning applications using multidata causal discovery,” pp. 1–13, 2023, doi: 10.1017/eds.2023.21.

N. Jia, X. Tian, Y. Zhang, and F. Wang, “Semi-Supervised Node Classification with Discriminable Squeeze Excitation Graph Convolutional Networks,” IEEE Access, vol. 8, pp. 148226–148236, 2020, doi: 10.1109/ACCESS.2020.3015838.

T. Liu et al., “Spatial Channel Attention for Deep Convolutional Neural Networks,” Mathematics, vol. 10, no. 10. p. 1750, 2022, doi: 10.3390/math10101750.

S. Albahli and M. Nawaz, “DCNet: DenseNet-77-based CornerNet model for the tomato plant leaf disease detection and classification,” Front. Plant Sci., vol. 13, no. September, pp. 1–19, 2022, doi: 10.3389/fpls.2022.957961.

E. Ding, Y. Cheng, C. Xiao, Z. Liu, and W. Yu, “Efficient Attention Mechanism for Dynamic Convolution in Lightweight Neural Network,” Applied Sciences, vol. 11, no. 7. p. 3111, 2021, doi: 10.3390/app11073111.

X. Zhang, Y. Mao, Q. Yang, and X. Zhang, “A Plant Leaf Disease Image Classification Method Integrating Capsule Network and Residual Network,” IEEE Access, vol. 12, no. February, pp. 44573–44585, 2024, doi: 10.1109/ACCESS.2024.3377230.

M. Hassanin, S. Anwar, I. Radwan, F. S. Khan, and A. Mian, “Visual attention methods in deep learning: An in-depth survey,” Inf. Fusion, vol. 108, p. 102417, 2024, doi: https://doi.org/10.1016/j.inffus.2024.102417.

S. Senthil Pandi, A. Senthilselvi, J. Gitanjali, K. ArivuSelvan, J. Gopal, and J. Vellingiri, “Rice plant disease classification using dilated convolutional neural network with global average pooling,” Ecol. Modell., vol. 474, p. 110166, 2022, doi: https://doi.org/10.1016/j.ecolmodel.2022.110166.

Z. Wang and L. I. Wang, “An Attention Approach for Dimensionality Reduction That Can Correct Feature Collection Skewness,” IEEE Access, vol. 10, no. November, pp. 117273–117280, 2022, doi: 10.1109/ACCESS.2022.3220245.

S. M. Yasir and H. Kim, “Attention-Based Deep Feature Aggregation Network for Skin Lesion Classification,” Electron., vol. 14, no. 12, 2025, doi: 10.3390/electronics14122364.

H. Zhu et al., “MS-HNN: Multi-Scale Hierarchical Neural Network With Squeeze and Excitation Block for Neonatal Sleep Staging Using a Single-Channel EEG,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 31, pp. 2195–2204, 2023, doi: 10.1109/TNSRE.2023.3266876.

X. Jin et al., “Delving deep into spatial pooling for squeeze-and-excitation networks,” Pattern Recognit., vol. 121, p. 108159, 2022, doi: 10.1016/J.PATCOG.2021.108159.

B. Chen, T. Zhao, J. Liu, and L. Lin, “Multipath feature recalibration DenseNet for image classification,” Int. J. Mach. Learn. Cybern., vol. 12, no. 3, pp. 651–660, 2021, doi: 10.1007/s13042-020-01194-4.

M. K. H. Hussein and O. N. Ucan, “3D Face Anti-Spoofing with Dense Squeeze and Excitation Network and Neighborhood-Aware Kernel Adaptation Scheme,” IEEE Access, vol. 13, no. February, pp. 43145–43167, 2025, doi: 10.1109/ACCESS.2025.3548418.

J. Zhang, J. Ding, J. Wang, Z. Zhang, J. Tan, and X. Ge, “Spatial heterogeneity response of soil salinization inversion cotton field expansion based on deep learning,” Front. Plant Sci., vol. 15, no. November, pp. 1–17, 2024, doi: 10.3389/fpls.2024.1437390.

X. Wang and W. Cao, “Bit-Plane and Correlation Spatial Attention Modules for Plant Disease Classification,” IEEE Access, vol. 11, no. July, pp. 93852–93863, 2023, doi: 10.1109/ACCESS.2023.3309925.

S. Zhong, W. Wen, and J. Qin, “SPEM: Self-adaptive Pooling Enhanced Attention Module for Image Recognition BT - MultiMedia Modeling,” 2023, pp. 41–53.

D. Lv, C. Zhao, H. Ye, Y. Fan, and X. Shu, “GS-YOLO: A Lightweight SAR Ship Detection Model Based on Enhanced GhostNetV2 and SE Attention Mechanism,” IEEE Access, vol. 12, no. July, pp. 108414–108424, 2024, doi: 10.1109/ACCESS.2024.3438797.

P. Chen, “An efficient mixed attention module,” no. February, pp. 496–507, 2023, doi: 10.1049/cvi2.12184.

B. M. Abuhayi and A. A. Mossa, “Coffee disease classification using Convolutional Neural Network based on feature concatenation,” Informatics Med. Unlocked, vol. 39, Jan. 2023, doi: 10.1016/j.imu.2023.101245.

L. Falaschetti, L. Manoni, D. Di Leo, D. Pau, V. Tomaselli, and C. Turchetti, “A CNN-based image detector for plant leaf diseases classification,” HardwareX, vol. 12, p. e00363, 2022, doi: 10.1016/j.ohx.2022.e00363.

M. T. Ahad, Y. Li, B. Song, and T. Bhuiyan, “Comparison of CNN-based deep learning architectures for rice diseases classification,” Artif. Intell. Agric., vol. 9, pp. 22–35, Sep. 2023, doi: 10.1016/j.aiia.2023.07.001.

L. M. Abouelmagd, M. Y. Shams, H. S. Marie, and A. E. Hassanien, “An optimized capsule neural networks for tomato leaf disease classification,” Eurasip J. Image Video Process., vol. 2024, no. 1, 2024, doi: 10.1186/s13640-023-00618-9.

K. H. Rhee, “Generation of Novelty Ground Truth Image Using Image Classification and Semantic Segmentation for Copy-Move Forgery Detection,” IEEE Access, vol. 10, pp. 2783–2796, 2022, doi: 10.1109/ACCESS.2021.3136781.

L. Alzubaidi et al., Review of deep learning: concepts, CNN architectures, challenges, applications, future directions, vol. 8, no. 1. Springer International Publishing, 2021.

I. Khan, S. S. Sohail, D. Ø. Madsen, and B. K. Khare, “Deep transfer learning for fine-grained maize leaf disease classification,” J. Agric. Food Res., vol. 16, no. March, 2024, doi: 10.1016/j.jafr.2024.101148.

J. Hua, J. Zou, J. Rao, H. Yin, and J. Chen, “ECG Signals Deep Compressive Sensing Framework Based on Multiscale Feature Fusion and SE Block,” IEEE Access, vol. 11, no. September, pp. 104359–104372, 2023, doi: 10.1109/ACCESS.2023.3316487.

M. Varan, J. Azimjonov, and B. Macal, “Enhancing Prostate Cancer Classification by Leveraging Key Radiomics Features and Using the Fine-Tuned Linear SVM Algorithm,” IEEE Access, vol. 11, no. August, pp. 88025–88039, 2023, doi: 10.1109/ACCESS.2023.3306515.

H. S. El-Assiouti, H. El-Saadawy, M. N. Al-Berry, and M. F. Tolba, “Lite-SRGAN and Lite-UNet: Toward Fast and Accurate Image Super-Resolution, Segmentation, and Localization for Plant Leaf Diseases,” IEEE Access, vol. 11, no. June, pp. 67498–67517, 2023, doi: 10.1109/ACCESS.2023.3289750.