Facial Emotion Recognition (FER) is a key technology for identifying emotions based on facial expressions, with applications in human-computer interaction, mental health monitoring, and customer analysis. This study presents the development of a real-time emotion recognition system using Convolutional Neural Networks (CNNs) and OpenCV, addressing challenges such as varying lighting and facial occlusions. The system, trained on the FER2013 dataset, achieved 85% accuracy in emotion classification, demonstrating high performance in detecting happiness, sadness, and surprise. The results highlight the system's effectiveness in real-time applications, offering potential for use in mental health and customer behavior analysis.

Traffic accidents are one of the leading causes of death worldwide, where drowsiness while driving is a significant factor that reduces driver alertness. This study develops a real-time driver drowsiness detection system using the Eye Aspect Ratio (EAR) method to avoid this. EAR calculates the ratio of the upper and lower eyelid distances to detect signs of drowsiness based on changes in eye shape. This system utilizes the OpenCV and Dlib libraries to identify faces and measure EAR, with a threshold of 0.25 as a warning trigger. If the EAR value drops below the threshold in several consecutive frames, the system automatically activates an alarm to increase driver alertness. With the advantages of cost efficiency and ease of implementation without additional hardware, this system is suitable for various types of vehicles. The results show that this system is effective in providing early warnings, thus helping to reduce the risk of accidents due to drowsiness.

The growing number of vehicles in major cities has posed significant challenges in parking lot management. Motorists often have difficulty finding empty parking slots quickly, which not only wastes time but also aggravates traffic congestion and increases air pollution. This research develops a Python-based smart parking system by utilizing the OpenCV library to detect the status of parking slots in real-time. The system uses a camera as the main sensor and processes the image using techniques such as grayscale, Gaussian blur, and adaptive threshold to identify the parking slot status, whether empty or occupied, with good accuracy. The parking slot coordinate data is stored in CSV format to ensure efficient data management. Experimental results with video recordings show that the system is able to operate well in various parking conditions. The system proved to be cost-effective and easy to implement, making it an ideal solution for parking managers who want to improve management efficiency without being burdened with high costs. This research offers a practical solution to help motorists and parking managers optimize parking space usage, reduce search time, and minimize negative impacts such as congestion and carbon emissions.

This study presents the development of an automatic passenger counting system for public buses using YOLOv8 based on Convolutional Neural Networks (CNN). The system detects and counts passengers in real-time to optimize bus capacity and enhance operational efficiency. Results indicate that the system achieves high accuracy in the front camera view (confidence score of 0.82). However, in the rear camera view the accuracy slightly decreases (confidence score of 0.76) due to object overlap, emphasizing the importance of proper camera placement. The system offers potential improvements in bus capacity management, reduced operational costs, and enhanced passenger comfort. These findings contribute to advancing smarter and more efficient public transportation systems.

Abstract. This study presents a web-based system for identifying traditional herbal leaves using K-Nearest Neighbors (KNN) and image processing techniques focused on analyzing leaf shape and color. The dataset used consists of images of various types of herbal leaves, providing a basis for classification and medicinal benefit information retrieval. The system was tested with multiple leaf samples to assess accuracy, speed, and effectiveness in identifying leaf types based on visual characteristics. Results show that the system can recognize different types of herbal leaves and display information on their medicinal properties in a user-friendly interface..

In recent years, virtual assistants have become an integral part of everyday life, simplifying routine tasks and allowing users to focus on more important matters. This research aiming to design GiggleGate, a virtual desktop assistant integrated with both face and speech recognition technology to enhance authentication security. The objective is to develop an authentication system that not only verifies user identity but also provides a more intuitive experience and seamless interaction. The research employs a development methodology to create and implement the system, which integrates face recognition via OpenCV and speech recognition via a Python library. The findings indicate that the integration of these technologies enhances security and user experience by offering dual-factor authentication. The system is expected to contribute to more secure and accessible virtual assistant applications, offering both a practical and efficient solution for users. The implications of this study suggest that the combination of face and speech recognition can provide an effective means to protect user privacy and improve the overall functionality of desktop assistants.

The development of technology with digital image processing is often utilized to solve various problems in image processing, such as facial recognition, object detection, and interaction between users. In this study, we developed an interactive hand gesture-based game titled "Slap Mosquito" that utilizes image processing techniques to control the game through hand gestures. Using Rapid Application Development (RAD), Python, OpenCV, and Pygame methodologies, this game allows users to slap mosquitoes virtually in real-time through hand gesture recognition that is read by the camera and translated into in-game actions. RAD allows rapid development iterations and improvements based on user feedback, which is essential for improving system responsiveness and accuracy. This study focuses on detection precision, system responsiveness, and the impact of lighting on game performance, as measured using frames per second (FPS) and user gameplay results. The test results show that optimal lighting meets high detection accuracy, while low lighting conditions have a negative impact on accuracy and responsiveness. The results of this study provide insights for further development of gesture-based applications, especially regarding the importance of optimizing technical parameters and RAD methodology in improving user experience.