Article: Revolutionising hyperspectral imaging: a hybrid deep learning and optimisation approach for superior classification Journal: International Journal of Signal and Imaging Systems Engineering (IJSISE) 2025 Vol.14 No.1 pp.58 - 75 Abstract: Accurate emotion detection from speech signals is essential for enhancing human-computer interaction (HCI) systems. However, existing SER methods suffer from poor feature representation and limited dataset diversity, resulting in suboptimal performance. To address these challenges, proposes an advanced deep bottleneck residual convolutional neural network (DBR-CNN) integrated with the SEResNeXt-101 feature extraction and optimised using the coati optimisation algorithm (COA). The model is trained and evaluated on four benchmark datasets: URDU, EMO-DB, EMOVO, and SAVEE. In the pre-processing phase, speech signals undergo noise reduction and normalisation to enhance data quality. The SEResNeXt-101 extractor then captures high-level features with reduced complexity, which are processed by the DBR-CNN to classify emotions with greater accuracy. The COA fine-tunes the model to improve classification efficiency. Experimental results demonstrate that the proposed model significantly outperforms state-of-the-art (SOTA) methods, achieving average accuracies of 99.43% on URDU, 99.625% on EMO-DB, 99.62% on EMOVO, and 99.66% on SAVEE datasets. Inderscience Publishers - linking academia, business and industry through research

Title: Revolutionising hyperspectral imaging: a hybrid deep learning and optimisation approach for superior classification

Authors: M. John Henry; Kambala Vijaya Kumar

Addresses: Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, 522302, Andhra Pradesh, India ' Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, 522302, Andhra Pradesh, India

Abstract: Accurate emotion detection from speech signals is essential for enhancing human-computer interaction (HCI) systems. However, existing SER methods suffer from poor feature representation and limited dataset diversity, resulting in suboptimal performance. To address these challenges, proposes an advanced deep bottleneck residual convolutional neural network (DBR-CNN) integrated with the SEResNeXt-101 feature extraction and optimised using the coati optimisation algorithm (COA). The model is trained and evaluated on four benchmark datasets: URDU, EMO-DB, EMOVO, and SAVEE. In the pre-processing phase, speech signals undergo noise reduction and normalisation to enhance data quality. The SEResNeXt-101 extractor then captures high-level features with reduced complexity, which are processed by the DBR-CNN to classify emotions with greater accuracy. The COA fine-tunes the model to improve classification efficiency. Experimental results demonstrate that the proposed model significantly outperforms state-of-the-art (SOTA) methods, achieving average accuracies of 99.43% on URDU, 99.625% on EMO-DB, 99.62% on EMOVO, and 99.66% on SAVEE datasets.

Keywords: HIS; hyperspectral image; IDRCNN; improved deep residual convolutional neural network; pre-processing; ConvNeXt; optimisation algorithm.

DOI: 10.1504/IJSISE.2025.150033

International Journal of Signal and Imaging Systems Engineering, 2025 Vol.14 No.1, pp.58 - 75

Received: 04 Mar 2025
Accepted: 15 Sep 2025
Published online: 21 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Revolutionising hyperspectral imaging: a hybrid deep learning and optimisation approach for superior classification

Keep up-to-date