Tea variety identification device and method based on LED array and convolutional neural network

Abstract

The invention relates to a tea variety identification device and method based on an LED array and a convolutional neural network. According to the invention, the LED array adopts a plurality of LEDs,wherein the central wavelength of the main LED ranges from 370 nm to 380 nm, the LED with the wavelength can excite fluorescence of tea polyphenol, flavone, anthocyanin and chlorophyll in tea leaves,useful information is provided to the maximum extent, the central wavelength of the auxiliary LED ranges from 360 nm to 500 nm, and the auxiliary LED is controlled by the calculation control unit in an alternate lightening mode. The mini spectrometer is used for collecting spectral data of tea to be identified and sending the spectral data to the calculation control unit. The calculation control unit converts the spectral data into two-dimensional data and sends the two-dimensional data to the trained convolutional neural network to obtain the variety of the tea to be identified. The device islight and portable, can be held by hand, and can be used for quickly and accurately identifying tea varieties.

Description

technical field [0001] The invention belongs to the technical field of optical sensing and analysis and detection instruments, and relates to a hand-held tea variety identification method and identification device based on a multi-wavelength LED array and a convolutional neural network. Background technique [0002] Tea is the most popular non-alcoholic beverage in the world, and it is grown and sold all over the world. According to the degree of fermentation, tea leaves are divided into green tea, black tea, green tea, white tea, yellow tea and dark tea. According to different types of tea, there are West Lake Longjing tea, Dongting Biluochun, Huangshan Maofeng, Lushan Yunwu tea, Lu'an Guapian, Junshan Yinzhen, Xinyang Maojian, Wuyi rock tea, Anxi Tieguanyin, Qimen black tea and other classifications. According to the different picking time, there are Qingming tea, Grain Rain tea, summer tea and autumn tea. Variety, season, origin, etc. will all have an impact on the qual...

AI Technical Summary

Problems solved by technology

Because the systems of these two schemes use the structure of the optical fiber to collect light to the micro spectrometer, the change of the bending state of the optical fiber will cause the change of the spectral curve during measurement, which will affect the measurement accuracy.

Images