Construction of Spatial Experience and Interaction Model of Industrial Heritage Based on Virtual Reality Technology

Abstract

The term "industrial heritage" describes the tangible remnants of a society's industrialization and economic growth, including buildings, locations, and artefacts. Various features such as factories, mills, mines, railroads, industrial machinery, warehouses, and other industrial infrastructure are all part of this history. These material remains serve as a testament to the changes that industrialization brought about in terms of technology, society, and the economy. In this manuscript, Construction of Spatial Experience and Interaction Model of Industrial Heritage Based on Virtual Reality Technology (BVR-SE-IMIH-TINN) are proposed. Initially input data is gathered from real time data is taken from camera or mobile phone. To execute this, input image is pre-processed using Generalized Multi-kernel Maximum Correntropy Kalman Filter (GMMCKF), it removes the noise from collected the data; then the Pre-processed data is feature extracted using Synchro spline-kernelled chirplet extracting transform (SSCET). In feature extraction SSCET is extract some feature such as Geometric features likes area, slope, centroids and perimeter. Then, the extracted data is fed to Thermodynamics-Informed Neural Network (TINN) for effectively categorize Interaction Model of Industrial Heritage. In general, TINN doesn’t express adapting optimization approaches to determine optimal parameters to ensure accurate virtual reality heritage picture collection. Hence, the Binary Arithmetic Optimization Algorithm (BAOA) to optimize Heterogeneous Thermodynamics-Informed Neural Network which accurately categorized the industrial heritage. Then the proposed BVR-SE-IMIH-TINN is implemented and the performance metrics such as Accuracy, Recall, Precision, F1- Score, Specificity, and Computation Time are analyzed. Performance of BVR-SE-IMIH-TINN approach attains 18.41%, 24.08% and 32.57% higher accuracy, 19.21%, 20.08% and 21.57% higher recall and attains 20.31%, 21.08% and 22.57% higher precision when analyzed with existing techniques likes intelligent splicing method of virtual reality Lingnan cultural heritage panorama depend on automatic machine learning (ISM-VLHP-AML), key technologies of digital protection of historical with cultural heritage depend on virtual reality technology (DPH-CH-VRT), Reconstruction of Industrial and Historical Heritage for Cultural Enrichment Utilizing Virtual  and Augmented Reality (RI-HCE-VAR) methods respectively.