Smart City Gnosys

Smart city article details

Title	Environmental Sound Classification With Low-Complexity Convolutional Neural Network Empowered By Sparse Salient Region Pooling
ID_Doc	24260
Authors	Seresht H.R.; Mohammadi K.
Year	2023
Published	IEEE Access, 11
DOI	http://dx.doi.org/10.1109/ACCESS.2022.3232807
Abstract	Environmental Sound Classification (ESC) is an important field in a broad range of applications, such as smart cities, audio surveillance, and health care. Recently, Convolutional Neural Networks (CNNs) have taken the lead from traditional approaches and have produced promising results. However, the achieved improvements are often accompanied by increasing depth, complexity, and size of the network, which prevents their usage in many practical applications. In this work, our goal is to empower a small-size low-complexity CNN model to achieve superior performance. To this end, we concentrate on the importance of global pooling technique, which is less investigated in ESC. In most previous works, models utilize global average pooling layer which does not consider regional saliency, and thus weakens the salient time-frequency regions contributions to the classification, and also to the training of convolutional kernels. We propose a novel global pooling method, called Sparse Salient Region Pooling (SSRP), which computes the channel descriptors using a sparse subset of features, and guides the model to effectively learn from the more salient time-frequency regions. Experimental results demonstrate that the proposed model with only 700K parameters yields accuracies of 86.7% on ESC-50 and 94.8% on ESC-10, which are comparable to that of the state-of-the-art methods. Compared to the baseline model, our model achieves absolute improvement of 21.8% in accuracy on ESC-50, with 98% smaller model size. Our visual analyses show that SSRP intensifies the responses of low-energy regions such that they contribute even more than high-energy regions to the classification of specific sound classes. © 2013 IEEE.
Author Keywords	Convolutional neural networks; environmental sound classification; global feature pooling; low complexity; regional saliency

Similar Articles

Id	Similarity	Authors	Title	Published
14555	0.938	Seker H.; Inik O.	Cnnsound: Convolutional Neural Networks For The Classification Of Environmental Sounds	ACM International Conference Proceeding Series (2020)
24259	0.907	Das J.K.; Chakrabarty A.; Piran M.J.	Environmental Sound Classification Using Convolution Neural Networks With Different Integrated Loss Functions	Expert Systems, 39, 5 (2022)
14548	0.895	İnik Ö.	Cnn Hyper-Parameter Optimization For Environmental Sound Classification	Applied Acoustics, 202 (2023)
14284	0.895	Reddy B.S.; Chowdary D.M.; Srinivas R.; Rahmani M.O.	Classification Of Environmental And Urban Sounds Using Deep Learning Techniques	4th IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics, ICDCECE 2025 (2025)
26145	0.888	Fang Z.; Yin B.; Du Z.; Huang X.	Fast Environmental Sound Classification Based On Resource Adaptive Convolutional Neural Network	Scientific Reports, 12, 1 (2022)
48795	0.887	Liu Z.; Yeh W.-C.	Simplified Swarm Optimisation For Cnn Hyperparameters: A Sound Classification Approach	International Journal of Web and Grid Services, 20, 1 (2024)
14305	0.885	Agarwal M.; Gill K.S.; Chattopadhyay S.; Singh M.	Classification Of Urban Sound Using Sequential Convolutional Neural Network (Cnn) Model And Its Visualisation	2024 IEEE International Conference on Information Technology, Electronics and Intelligent Communication Systems, ICITEICS 2024 (2024)
60186	0.875	Agarwal M.; Gill K.S.; Aggarwal P.; Rawat R.S.; Sunil G.	Urban Sound Classification Using Vgg19 Convolutional Neural Network (Cnn) Model And Its Visualisation	4th International Conference on Innovative Practices in Technology and Management 2024, ICIPTM 2024 (2024)
58279	0.874	Vijay M.; Ruthwik Saran K.; Reddy K.R.; Aditya Ram K.; Babu J.Y.	Towards Robust Environmental Sound Classification: A Deep Learning Approach Leveraging Time-Frequency Representations	2nd International Conference on Emerging Research in Computational Science, ICERCS 2024 (2024)
24672	0.871	Lamrini M.; Chkouri M.Y.; Touhafi A.	Evaluating The Performance Of Pre-Trained Convolutional Neural Network For Audio Classification On Embedded Systems For Anomaly Detection In Smart Cities	Sensors, 23, 13 (2023)