Title: |
A Study on the Standard Model of Smart Street Light |
Authors: |
Jaemin Kim, Sangmuk Cho, Byunggoo Lee, Jinsheon Kim |
Source: |
International Journal of Latest Engineering Research and Applications, pp 01 - 10, Vol 09 - No. 04, 2024 |
Abstract: |
Smart streetlights are innovative solutions that emphasize energy efficiency and environmental friendliness. Through the transition to LED lighting, they achieve energy savings and have the potential for efficient facility management. The development of smart streetlight systems requires interoperability, compatibility, scalability, and adaptability. A smart streetlight system integrating lighting control, remote monitoring, and environmental sensing capabilities is needed, and various configurations are compared considering laws and regulations. A survey of the application status of smart streetlights domestically and internationally reveals that they offer various services and functions such as energy savings, remote control, sensor integration, and smart city applications. Successful implementations of smart streetlights can be observed through case studies in domestic cities. The system configuration and functional analysis of smart streetlights explain the roles of IoT platforms, network infrastructure, and streetlight controllers, as well as key technologies such as sensor-based road condition detection, lighting control, wireless communication, and power supply management. The challenges and prospects of smart streetlights cover considerations such as standardization between products, technological stabilization, cost issues, patent rights, and maintenance. It emphasizes the need for collaboration between governments and businesses and a comprehensive approach. This research highlights the potential of smart streetlights and the importance of improving urban lighting infrastructure. It will be helpful in supporting the application and dissemination of smart streetlights. |
Keywords: |
Smart Street Lighting; Dimming Control; Smart cities; Smart Lighting |
 |
Download Full Article |
DOI: |
10.56581/IJLERA.9.4.01-10 |
Title: |
Using Real-Time Detection Transformer in Park Management |
Authors: |
Tai-Tien Chen, Yao-Chung Chen, Chen-Yu Hao, Tien-Yin Chou, Ting-Hsuan Lai |
Source: |
International Journal of Latest Engineering Research and Applications, pp 11 - 17, Vol 09 - No. 04, 2024 |
Abstract: |
Object detection is one of the primary tasks in computer vision and a crucial tool for digital governance in smart cities. The YOLO (You Only Look Once) series of object detection techniques have been regarded as efficient methods. With the rise of Transformers, research in various tasks has stirred waves in the field of computer vision. RTDETR (Real Time Detection Transformer) emerges as a derivative method in the object detection domain. It is well known that each detection method has its own characteristics, and selecting an appropriate model according to task requirements is a common practice for executing the task. Therefore, this study retrained YOLOv8-x, YOLOv8-l, RTDETR-x, and RTDETR-l models separately for object categories required by our park management. The results show that in terms of detection accuracy, the YOLOv8-x and RTDETR-x models outperform the YOLOv8-l and RTDETR-l models, and the RTDETR-x demonstrates the best object detection capability. |
Keywords: |
Object detection, You Only Look Once, Real Time Detection Transformer |
|
Download Full Article |
DOI: |
10.56581/IJLERA.9.4.11-17 |
Title: |
Genre Classification and Musical Features Analysis |
Authors: |
Ozlem Kilickaya |
Source: |
International Journal of Latest Engineering Research and Applications, pp 18 - 33, Vol 09 - No. 04, 2024 |
Abstract: |
Genre classification in music is a fundamental task in machine learning research, with implications for music recommendation systems, content organization, and music analysis. This study delves into the process of genre classification and analysis of musical features extracted from audio files. Leveraging the GTZAN dataset, a widely used resource in music genre recognition research, this study demonstrates the extraction and exploration of audio features, including but not limited to spectral centroid, chroma features, and tempo. Three machine learning models—Support Vector Machines (SVM), Random Forest, and XGBoost—are employed for genre classification, with meticulous attention given to hyperparameter tuning and proper train/test set partitioning to prevent data leakage. Notably, precautions are taken to ensure that excerpts from the same song do not overlap between training and testing sets, a common pitfall observed in similar studies. By looking closely at misclassified genre pairs and giving detailed explanations of classification results, the study explores the subtleties of genre labeling and questions how humans detect musical traits are different between genres. This research contributes to advancing understanding in music genre classification and underscores the importance of rigorous methodology in machine learning-based music analysis. |
Keywords: |
Genre classification, Music analysis, Audio feature extraction, Machine learning, Hyperparameter tuning, Genre recognition, Music recommendation systems. |
|
Download Full Article |