Crowd monitoring and analysis has become increasingly used for unmanned aerial vehicle applications. From preventing stampede in high concentration crowds to estimating crowd density and to surveilling crowd movements, crowd monitoring and analysis have long been employed in the past by authorities and regulatory bodies to tackle challenges posed by large crowds. Conventional methods of crowd analysis using static cameras are limited due to their low coverage area and non-flexible perspectives and features. Unmanned aerial vehicles have tremendously increased the quality of images obtained for crowd analysis reasons, relieving the relevant authorities of the venues’ inadequacies and of concerns of inaccessible locations and situation. This paper reviews existing literature sources regarding the use of aerial vehicles for crowd monitoring and analysis purposes. Vehicle specifications, onboard sensors, power management, and an analysis algorithm are critically reviewed and discussed. In addition, ethical and privacy issues surrounding the use of this technology are presented.

Road anomalies and irregularities such as potholes and uneven surfaces are a common hazard in South East Asia and developing countries. Such hazards pose a threat to the safety and well-being of both civilians going about their daily routine and tourists who are exploring the city. Since bicycles and rickshaws are still a common mode of transport used by both civilians and tourists in many South East Asian countries, it is essential to improve the overall quality and smoothness of pavements which are traversed by these vehicles. Management of international sporting and recreational events also require satisfactory road and pavement conditions. Before pavement conditions can be improved, it is an essential prerequisite to obtain comprehensive information about road irregularities such as the location and also severity of the road irregularity (depth of the potholes and height of bumps). In this paper, we propose a method for obtaining mathematical models that represent the overall condition of the pavements that are part of a commonly traversed cycling route. Such mathematical models and coefficients can be stored in the cloud of an Internet of Things (IOT) data analytics systems subsequently leading to identification of regions with severe road irregularities. ABSTRAK: Kerosakan pada permukaan jalan raya merupakan salah satu faktor risiko kemalangan yang berlaku secara meluas di negara Asia Tenggara dan negara membangun yang lain. Memandangkan kenderaan seperti beca dan basikal masih diguna pakai secara meluas di negara membangun, adalah mustahak untuk membaik pulih kerosakan jalan raya. Pengurusan sukan antarabangsa dan rekreasi juga memerlukan keadaan jalan dan laluan pejalan kaki yang baik. Sebelum kerja membaik pulih dapat dilakukan, maklumat lengkap mengenai tahap kerosakan jalan raya dan lokasi kerosakan diperlukan. Dalam kajian ini satu kaedah telah diperkenalkan untuk mendapatkan persamaan matematik yang menggambarkan keadaan sebenar permukaan jalan raya, di mana sebahagiannya merupakan laluan berbasikal yang selalu digunakan. Model matematik dan pekali ini boleh di simpan dalam sistem analisis data awan Internet Benda (IOT) kemudiannya dapat mengenal pasti kawasan jalan yang rosak dan tidak rata.

Animal intrusion has caused many issues for human beings, especially monkeys. Monkeys have caused many problems such as yield crop damage, damage to human facilities and assets and stealing food. This study aims to investigate the use of deep learning to detect monkey presence accurately and use a proper repellent system to scare them away. A deep learning algorithm is constructed with supervised learning, which includes the monkey dataset with appropriate labelling of the object of interest. The detection of the monkey comes with a bounding box with respective confidence of detection. The result is then used to evaluate the accuracy of monkey detection. The accuracy of the trained model is assessed by evaluating its performance under varying conditions of camera quality and distances. The study focuses on proving the reliability of deep learning to detect monkeys and automatically perform corresponding actions like repelling monkeys. Hence this may reduce the reliance of manpower to secure a large space and improve safety issues.

Background: Stroke often results in upper limb impairment, requiring repetitive rehabilitation exercises that can be demotivating over time. Aim: This study aims to develop and evaluate a rehabilitation system called Catch & Heal , which combines haptic-based gameplay and social robot-mediated motivational support to enhance upper limb therapy and user experience. The novelty of this system lies in its unified design, which combines serious gaming, physical feedback, and social interaction, an integration that remains underdeveloped in current research. Method: Developed in Unity, the game adapts fishing mechanics to promote therapeutic wrist movements like push, pull, and rotation, mapped through joystick-based controls. Realistic haptic feedback, delivered via Arduino-driven vibration motors, simulates fish resistance to enhance motor engagement and proprioceptive response. A socially assistive robot provides real-time motivational cues, fostering emotional connection and sustained engagement. A feasibility study involving thirty healthy participants ( M = 22.73 years) in a controlled lab setting evaluated usability, engagement, emotional response, and perceptions of robot interaction. Results: Outcomes showed high satisfaction with the user interface ( M = 4.56), strong in-game immersion ( M = 3.56), and favorable reception of the robot’s support ( M = 4.35). Significant correlations were found between user interface simplicity and aesthetics ( r = 0.878), and between immersion and emotional outcomes ( r = 0.872), highlighting therapeutic potential. Conclusions: This integrated approach offers a promising direction for engaging and emotionally responsive rehabilitation, with future refinement focused on stroke patients.