International Journal of Neutrosophic Science IJNS 2690-6805 2692-6148 10.54216/IJNS https://www.americaspg.com/journals/show/3128 2020 2020 Department of Computer Science, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 21589, Saudi Arabia; Center of Research Excellence in Artificial Intelligence and Data Science, King Abdulaziz University, Jeddah 21589, Saudi Arabia Eaman Eaman Neutrosophic set (NS) and Neutrosophic logic (NL) play a major part in approximation theory. They are generalizations of intuitionistic fuzzy sets and logic correspondingly. Rough NS (RNS) combines the concepts of RS and NL to deal with vagueness, uncertainty, and imprecision in information. By integrating truth, indeterminacy, and false degrees, RNS provides a more solid basis for analyzing and classifying complicated data. Particularly, this makes it powerful in applications where incompleteness and ambiguity of data are ubiquitous. Smart cities are a current trend to contain information and communication technologies (ICTs) in the progression of great urban cities. It would be beneficial in defining the city's movement by monitoring the regular flow of traffic jams and visitors. One important characteristic of smart cities is Crowd management, which assists in safety and enjoyable experiences for the residents and visitors. Since the crowd density (CD) classification method encounters tasks including inter-scene, non-uniform density, and intra-scene deviations, occlusion and convolutional neural networks (CNNs) approaches were beneficial. This work focuses on the design of Automated Crowd Density Recognition using the Rough Neutrosophic Set for Smart Cities (ACDR-RNSSC) method in urban management. The presented ACDR-RNSSC method focuses on identifying various types of crowd densities in smart cities. Firstly, the ACDR-RNSSC method utilizes the ResNet50 method for feature extraction. Second, the classification is done using RNS. RNS is utilized for its ability to manage the vagueness and uncertainty in crowd density statistics. Lastly, the parameter is fine-tuned using the Fruit Fly Optimization Algorithm (FOA). This ensures that the model attains high robustness and accuracy in forecasting crowd density. The empirical analysis of the ACDR-RNSSC method is examined under benchmark crowd dataset and the outcomes are tested using various metrics. This study states the improvement of the ACDR-RNSSC method over existing techniques. 2025 2025 117 128 10.54216/IJNS.250210 https://www.americaspg.com/articleinfo/21/show/3128