Research on concentration detection algorithm of embedded gas detection system based on deep learning

In daily life, whether it is the house where people live or the car they drive, air pollution can be seen almost everywhere, and human life and health are seriously threatened. There is an urgent need for a detection system for fast and accurate detection and assessment of ambient air quality. On the other hand, in the medical field, the detection of human exhaled gas can realize non-invasive and rapid early disease screening, but to achieve this purpose, a fast, accurate, highly sensitive and selective gas detection system is also required. At present, the research and development of gas detection system has attracted more and more attention of scholars, and the gas detection technology using porphyrin sensor array (PSA) chip is a hot topic in this research direction in recent years. Compared with traditional methods, this method has many obvious advantages and can meet the needs of fast, accurate and sensitive detection. In this paper, a PSA chip is used as a gas sensor, a set of PSA image acquisition and processing software is designed and developed on the Linux platform, and successfully transplanted to the ARM platform to build a complete gas detection system. Next, the data was collected and processed with ammonia gas as the detection object, and the quantitative analysis model of gas concentration was studied.