Muhammad Asif Khan¹, Dildar Hussain², Bhuyan Kaibalya Prasad³, Irfan Ullah⁴, Inayat Khan⁵, Jawad Khan^6,*, Yeong Hyeon Gu^2,*, Pavlos Kefalas⁷

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5451-5468, 2025, DOI:10.32604/cmc.2025.069134 - 23 October 2025

Abstract Dialogue State Tracking (DST) is a critical component of task-oriented spoken dialogue systems (SDS), tasked with maintaining an accurate representation of the conversational state by predicting slots and their corresponding values. Recent advances leverage Large Language Models (LLMs) with prompt-based tuning to improve tracking accuracy and efficiency. However, these approaches often incur substantial computational and memory overheads and typically address slot extraction implicitly within prompts, without explicitly modeling the complex dependencies between slots and values. In this work, we propose PUGG, a novel DST framework that constructs schema-driven prompts to fine-tune GPT-2 and utilizes its tokenizer… More >

Sahbi Boubaker^1,2,*, Salim Hadj Said³, Souad Kamel¹, Habib Dimassi³

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5703-5721, 2025, DOI:10.32604/cmc.2025.069047 - 23 October 2025

Abstract Accurate trajectory tracking in lower-limb exoskeletons is challenged by the nonlinear, time-varying dynamics of human-robot interaction, limited sensor availability, and unknown external disturbances. This study proposes a novel control strategy that combines flatness-based control with two cascaded observers: a high-gain observer to estimate unmeasured joint velocities, and a nonlinear disturbance observer to reconstruct external torque disturbances in real time. These estimates are integrated into the control law to enable robust, state-feedback-based trajectory tracking. The approach is validated through simulation scenarios involving partial state measurements and abrupt external torque perturbations, reflecting realistic rehabilitation conditions. Results confirm More >

Sheeba Razzaq^1,*, Majid Iqbal Khan²

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5377-5398, 2025, DOI:10.32604/cmc.2025.068539 - 23 October 2025

Abstract Multi-Object Tracking (MOT) represents a fundamental but computationally demanding task in computer vision, with particular challenges arising in occluded and densely populated environments. While contemporary tracking systems have demonstrated considerable progress, persistent limitations—notably frequent occlusion-induced identity switches and tracking inaccuracies—continue to impede reliable real-world deployment. This work introduces an advanced tracking framework that enhances association robustness through a two-stage matching paradigm combining spatial and appearance features. Proposed framework employs: (1) a Height Modulated and Scale Adaptive Spatial Intersection-over-Union (HMSIoU) metric for improved spatial correspondence estimation across variable object scales and partial occlusions; (2) a feature More >

Zeeshan Ali¹, Jihoon Moon², Saira Gillani³, Sitara Afzal⁴, Maryam Bukhari⁵, Seungmin Rho^6,*

CMES-Computer Modeling in Engineering & Sciences, Vol.144, No.2, pp. 2263-2286, 2025, DOI:10.32604/cmes.2025.067743 - 31 August 2025

Abstract Surveillance systems can take various forms, but gait-based surveillance is emerging as a powerful approach due to its ability to identify individuals without requiring their cooperation. In the existing studies, several approaches have been suggested for gait recognition; nevertheless, the performance of existing systems is often degraded in real-world conditions due to covariate factors such as occlusions, clothing changes, walking speed, and varying camera viewpoints. Furthermore, most existing research focuses on single-person gait recognition; however, counting, tracking, detecting, and recognizing individuals in dual-subject settings with occlusions remains a challenging task. Therefore, this research proposed a… More >

Shitong Ye¹, Bo Yang^2,*, Hao Quan³, Shan Liu⁴, Minyi Tang⁵, Jiawei Tian^6,*

CMES-Computer Modeling in Engineering & Sciences, Vol.144, No.2, pp. 2287-2306, 2025, DOI:10.32604/cmes.2025.066195 - 31 August 2025

Abstract Computer-aided surgical navigation technology helps and guides doctors to complete the operation smoothly, which simulates the whole surgical environment with computer technology, and then visualizes the whole operation link in three dimensions. At present, common image-guided surgical techniques such as computed tomography (CT) and X-ray imaging (X-ray) will cause radiation damage to the human body during the imaging process. To address this, we propose a novel Extended Kalman filter-based model that tracks the puncture needle-point using an ultrasound probe. To address the limitations of Kalman filtering methods based on position and velocity, our method of More >

Pallabi Biswas^1,#, Samarendra Nath Sur^2,#,*, Rabindranath Bera³, Agbotiname Lucky Imoize⁴, Chun-Ta Li^5,*

CMES-Computer Modeling in Engineering & Sciences, Vol.144, No.1, pp. 83-146, 2025, DOI:10.32604/cmes.2025.067724 - 31 July 2025

Abstract Automotive radar has emerged as a critical component in Advanced Driver Assistance Systems (ADAS) and autonomous driving, enabling robust environmental perception through precise range-Doppler and angular measurements. It plays a pivotal role in enhancing road safety by supporting accurate detection and localization of surrounding objects. However, real-world deployment of automotive radar faces significant challenges, including mutual interference among radar units and dense clutter due to multiple dynamic targets, which demand advanced signal processing solutions beyond conventional methodologies. This paper presents a comprehensive review of traditional signal processing techniques and recent advancements specifically designed to address… More > Graphic Abstract

Yu-Shiuan Tsai^*, Yuk-Hang Sit

CMES-Computer Modeling in Engineering & Sciences, Vol.143, No.3, pp. 3065-3090, 2025, DOI:10.32604/cmes.2025.064783 - 30 June 2025

Abstract To improve small object detection and trajectory estimation from an aerial moving perspective, we propose the Aerial View Attention-PRB (AVA-PRB) model. AVA-PRB integrates two attention mechanisms—Coordinate Attention (CA) and the Convolutional Block Attention Module (CBAM)—to enhance detection accuracy. Additionally, Shape-IoU is employed as the loss function to refine localization precision. Our model further incorporates an adaptive feature fusion mechanism, which optimizes multi-scale object representation, ensuring robust tracking in complex aerial environments. We evaluate the performance of AVA-PRB on two benchmark datasets: Aerial Person Detection and VisDrone2019-Det. The model achieves 60.9% mAP@0.5 on the Aerial Person… More >

Yi-Chen Lee¹, Yann-Horng Lin², Wen-Jer Chang^2,*, Muhammad Shamrooz Aslam^3,*, Zi-Yao Lin²

CMES-Computer Modeling in Engineering & Sciences, Vol.143, No.2, pp. 1433-1461, 2025, DOI:10.32604/cmes.2025.064717 - 30 May 2025

Abstract An optimal fuzzy tracking synthesis for nonlinear discrete-time descriptor systems is discussed through the Parallel Distributed Compensation (PDC) approach and the Proportional-Difference (P-D) feedback framework. Based on the Takagi-Sugeno Fuzzy Descriptor Model (T-SFDM), a nonlinear discrete-time descriptor system is represented as several linear fuzzy subsystems, which facilitates the linear P-D feedback technique and streamlines the fuzzy controller design process. Leveraging the P-D feedback fuzzy controller, the closed-loop T-SFDM can be transformed into a standard system that guarantees non-impulsiveness and causality for the nonlinear discrete-time descriptor system. In view of the disturbance problems, a passive performance… More >

He Su, Jianwei Gao, Kang Kong^*

CMC-Computers, Materials & Continua, Vol.83, No.3, pp. 4193-4213, 2025, DOI:10.32604/cmc.2025.063355 - 19 May 2025

Abstract Accurate and robust navigation in complex surgical environments is crucial for bronchoscopic surgeries. This study purposes a bronchoscopic lumen feature matching network (BLFM-Net) based on deep learning to address the challenges of image noise, anatomical complexity, and the stringent real-time requirements. The BLFM-Net enhances bronchoscopic image processing by integrating several functional modules. The FFA-Net preprocessing module mitigates image fogging and improves visual clarity for subsequent processing. The feature extraction module derives multi-dimensional features, such as centroids, area, and shape descriptors, from dehazed images. The Faster R-CNN Object detection module detects bronchial regions of interest and… More >

Praveen Kumar Sekharamantry^1,2,*, Farid Melgani¹, Roberto Delfiore³, Stefano Lusardi³

CMC-Computers, Materials & Continua, Vol.83, No.3, pp. 4215-4238, 2025, DOI:10.32604/cmc.2025.062686 - 19 May 2025

Abstract Recent advances in computer vision and artificial intelligence (AI) have made real-time people counting systems extremely reliable, with experts in crowd control, occupancy supervision, and security. To improve the accuracy of people counting at entry and exit points, the current study proposes a deep learning model that combines You Only Look Once (YOLOv8) for object detection, ByteTrack for multi-object tracking, and a unique method for vector-based movement analysis. The system determines if a person has entered or exited by analyzing their movement concerning a predetermined boundary line. Two different logical strategies are used to record… More >