Comparative Study of Different Methods for Fire Detection Using Convolutional Neural Network (CNN)

Rachana, P; Rajalakshmi, B.; Bhat, Tushar; Kaur, Sukhmanjeet; Bimali, Stuti

doi:10.1109/icssit53264.2022.9716284

Cited by 5 publications

(1 citation statement)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These metrics provide insight into the system's ability to correctly identify and classify objects and scenes, which is crucial for generating relevant and precise descriptions for users. The system's responsiveness, a measure of its efficiency, is evaluated by the latency between user input and system output [33]. This is critical for real-time applications, particularly for sign language translation in videos, where delays can disrupt the user's understanding and engagement with the content.…”

Section: Implementation and Performance Evaluationmentioning

confidence: 99%

A Review on Enhancing Accessibility Through Image and Video Processing: Solutions for Differently Abled Individuals

Anjimoon,

Sobti

et al. 2024

E3S Web Conf.

View full text Add to dashboard Cite

This paper presents innovative methodologies in image and video processing aimed at augmenting accessibility for differently abled individuals. Central to this research is the development of advanced algorithms that enable enhanced interpretation and interaction with multimedia content, thereby empowering users with sensory impairments. The study introduces a multi-layered framework that integrates adaptive filtering, object recognition, and augmented reality, tailored to the needs of users with visual and auditory challenges. Semantic scene analysis is leveraged to provide descriptive audio annotations for the visually impaired, facilitating a comprehensive understanding of visual data. For individuals with hearing impairments, the system incorporates real-time sign language interpretation within videos, utilizing deep learning techniques. The efficacy of these solutions is measured against conventional accessibility tools, demonstrating significant improvements in user engagement and comprehension. A novel contribution of this research is the application of machine learning to calibrate the system according to individual user profiles, ensuring a personalized and intuitive user experience. The scalability of the proposed system is validated through its implementation across various platforms and content formats. The findings suggest that such technological advancements have the potential to significantly reduce the barriers faced by differently abled individuals in accessing multimedia information.

show abstract

Section: Implementation and Performance Evaluationmentioning

confidence: 99%