Dangerous chemical rapid detection method based on neural network architecture search

• Main Authors: LIU YINGLI, CHEN SHAOHUA, ZHANG XUESHENG, MOU TAOTAO
• Format: Patent
• Language: Chinese; English
• Published: 20.09.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

The invention discloses a dangerous chemical rapid detection method based on neural network architecture search. The method comprises the following steps: step 1, acquiring spectral data of dangerous chemicals; step 2, network design of a neural network architecture search method based on AutoKeras; step 3, obtaining an optimal network architecture; in the step 1, for solid chemicals, laser is directly irradiated on the surface of a sample at a focal length of 7.5 ms to collect spectral data, and for liquid chemicals, the sample needs to be placed in a standard sample bottle to collect spectral data; the spectrum of the dangerous chemicals is measured through the Raman spectrometer, a neural network architecture search method is introduced, an AutoKeras framework is adopted to develop and design a self-network architecture, and compared with other neural network methods, the optimal network architecture is found through search and integration of network modules, performance evaluation and feedback, so that th