This study investigated the potential of using hyperspectral imaging working in visible and short-wave near infrared region (380-1030 nm) for rapid and non-invasive determination of the total flavonoid in mulberry fruit. Mulberry fruit with its sweet flavor is widely used in jam, pies, tarts, wines, and liquor, and is a delicacy among humans and birds alike. The quality evaluation of mulberry is usually determined by chemical or sensory analysis. However these methods are not capable of fast determination and real-time monitoring. Some of them are destructive, some take a long time to obtain the testing result for one fruit from the preparation to the end, and some need professional operators to finish the analysis. With the combination of the main advantages of spectroscopy and computer vision, hyperspectral imaging technique can simultaneously acquire spectral and spatial information in one system. Such ability let hyperspectral imaging be able to determine the inherent chemical and physical properties of the specimen as well as their spatial distribution simultaneously, which is critical for the quality prediction of agricultural and food products in a detailed way. In this work, a hyperspectral image of each fruit was acquired by using a pushbroom line-scanning HSI instrument. The system mainly consisted of an ImSpectorV17E imaging spectrograph covering the spectral range of 850-1750 nm, a high-resolution single piece digital camera, a camera lens, a specially assembled light unit consists of two 150 w quartz tungsten halogen lamps as the light source, and a conveyer belt operated by a stepper motor. Each fruit was placed on the moving table and then was scanned line by line to build a hyperspectral image (R₀) called ‘hypercube’ with a dimension of (x, y, λ). A 2-D image (y, λ) with the whole spectral dimension (λ) with one spatial dimension (y) was acquired at a time. A complete hyperspectral cube was taken as the line was scanned along the direction of x dimension, and was stored in a band-interleaved-by-line (BIL) format. Average spectra of each fruit were extracted from its hyperspectral image using the Region of Interests Function (ROI) of ENVI v4.6 software. Partial least-squares regression (PLSR) was used to relate the extracted spectral data and the total flavonoid content. PLSR can decompose both the spectral (independent variables) and concentration (dependent variables) information simultaneously, resulting in extracting a set of orthogonal factors called latent variables (LVs). In the decomposition process, dependent variables are actively considered in estimating the LVs to ensure that the first several LVs are most related for predicting dependent variables. Results showed that a good correlation was obtained between the reference total flavonoid content and spectral information. Due to our knowledge, it is the first time to determine the total flavonoid content of mulberry fruit and to generate the content distribution within the fruit using hyperspectral imaging.