In recent years, significant progress has been made in Face Recognition (FR) systems, finding diverse applications such as security, authentication, surveillance, and user convenience. Despite these advancements, a persistent challenge in FR systems is the occurrence of false positives, where individuals are mistakenly identified as matches. This issue can lead to security breaches, privacy concerns, and user dissatisfaction. Researchers are actively developing deep-learning-based algorithms to enhance FR systems by addressing false positive (FP) rates. This paper introduces an innovative approach to tackle false positives in real-time face recognition systems that operate on image streams using a filtering method. Instead of treating every identification as a match, our method dynamically filters false positive results using a larger database containing recent and necessary face images. This filtering technique significantly decreases the number of false positive cases, resulting in a more accurate and reliable face recognition system. The experimental results presented in this paper illustrate the effectiveness of our approach in reducing false positives across various challenging scenarios. By adapting to the specific context in which face recognition is applied, our method achieves a noteworthy reduction in false positives while maintaining a high level of accuracy and efficiency. In summary, mitigating false positives in face recognition stands as a crucial stride in unlocking the complete potential of this technology, simultaneously addressing apprehensions regarding its potential misuse. Our innovative false positive filtering approach presents a promising resolution to this challenge, laying the foundation for the development of more secure and reliable face recognition systems across diverse domains.