Portable in vivo measurement of apple sugar content based on mobile phone

With the development of science and technology, people's living standards continue to improve. Sugar content has been widely studied as an important index to measure fruit quality, but at present, most sugar content detection needs expensive equipment or accessories, which is difficult to enter daily life. In this paper, we propose a convenient scheme for detecting apple sugar degree based on multispectral and machine learning. With the mobile phone screen as the main light source, the front camera captures Apple pictures, and changes the color of the mobile phone screen to change the wavelength of the light source. By photographing different surfaces of apples under different wavelengths of visible light, obtain pictures of the same apple with different spectra, sort out the data, make the data set and train the machine learning network model, deploy the trained network into the app, and then predict the sugar value of apples, so as to achieve the purpose of rapid and nondestructive detection of apple sugar.