Non-destructive Measurement of Apple Sugar Content Based on Capsule Net

Traditional methods for testing apple sugar content and acidity are mostly destructive, affecting sales and consumption. At present, there is a lack of low-cost and intelligent detectors to meet people's requirements for apple quality testing. Infrared image detection is highly efficient and non-destructive. Quantitatively predict the sugar content of apples. Near-infrared light is an electromagnetic wave with a wavelength range of 780nm to 2526nm. The sugar in apples contains functional groups that easily absorb the near-infrared spectrum. Apples with different sugar levels have different spectral absorptions due to their functional groups. When the light source illuminates the surface of the apple, it is inside the apple. Diffuse transmission will occur. According to the different intensity of the diffuse transmitted light signal received by the detector, a stable objective function can be obtained by establishing the mathematical relationship between the received diffuse transmitted light signal and the sugar content of the apple. This paper uses literature research methods, empirical research methods, and quantitative analysis methods to process the apple infrared image algorithmically, and extract the information related to sugar content to establish a model to provide an idea for infrared non-destructive testing of apples. After experimental verification, the accuracy has reached very high.