The prediction system of bruising depth in guava based on storage operation was studied using a system Vis-NIR hyperspectral imaging in the visible and near-infrared (400-1000) nm ranges, which could develop a nondestructive method for predicting the bruise depth of guava. The fruit surface provided additional information not only about the position of the bruised area but also about its depth. Spectral images were acquired for 33 guavas over a period of four days after bruising, using a push-broom Vis-NIR hyperspectral imaging system. Individual guavas were then subjected to impact test by a steel ball at one of the levels height of impact test, i.e., 200 and 500mm. The hardware of the proposed system consists of a set of the workbench, controllable slider, two halogen light sources (150Watt) and a hyperspectral camera that is connected to PC via Camera Link. The software of system consists of reflectance image profile measurement, feature extraction, feature selection on spectral and spatial data, and bruise depth prediction model. The partial least squares regression (PLSR) models were used to develop prediction models on full wavelengths spectral data. The prediction model is used to get value prediction of bruising depth. The predicted results compared with the reference measurement result of bruising depth which obtained using optical properties. Better predictions of bruise depth were obtained from the PLS models for each impact height level, with the correlation coefficient of prediction or R 0,03 and root mean square error of prediction or RMSE 0,99%. This paper demonstrated that there is a feasibility of implementing hyperspectral imaging technique on the nondestructive bruise depth prediction of guava and suitable in an industrial sorting system for fruit quality, which would be useful for postharvest handling of fruit.