<![CDATA[This study aims to develop a web-based monitoring system utilizing computer vision and the Histogram of Oriented Gradients (HOG) algorithm to detect and count the number of people in a room. The research was motivated by the need for automated systems to replace manual methods, particularly during the COVID-19 pandemic, where contactless attendance tracking is critical. The HOG algorithm extracts visual features as image gradients representing human shapes, which are then classified using a machine learning model. The research methodology involved testing various camera configurations and angles, such as above door placement and room corners, with 80 and 60-degree camera angles. The system was built using Python, OpenCV for image processing, and Flask for web interface development. Data was collected through 200 trials, capturing live images with a camera. The results showed that the camera positioned above the door at an 80-degree angle achieved the highest detection accuracy at 66%, while other configurations showed reduced accuracy, averaging around 52%. The system also logs data, including time, detected human count, and captured images for further analysis. The findings highlight the importance of camera placement and angle in ensuring effective human detection. These results support the development of computer vision-based systems for practical applications such as crowd management, attendance counting, and public space monitoring. Future research is proposed to improve detection accuracy by integrating facial recognition technology and developing a more intuitive user interface. This study contributes substantially to improving the efficiency of automated systems for counting individuals in enclosed spaces, addressing various industrial and social needs.]]>
