Optimizing Air Quality Management with an Energy-Efficient Deep Learning Soft Sensor

Abstract

This paper presents a novel approach for monitoring air quality by utilizing Cryptogams, a bio-indicator that can accurately reflect pollution levels. We introduce an advanced and energy-efficient deformable active contour model designed to track the growth of transplanted Cryptogams across various pollution sites. Our study focuses on monitoring the vegetative development of Cryptogams over a span of two weeks, showcasing the effectiveness of our proposed energy-efficient contour tracing model in precise tracking, resulting in more reliable pollution monitoring.