Modeling of fruit morphological formation in melon is important for realizing virtual and digital plant growth. The objective of this study was to characterize the changes in patterns of fruit growth characters during plant development. In cultivar experiments, a high-resolution wireless vision sensor network has been developed to realize non-contact automatic uninterrupted measurement of the fruit shape micro-change (fruit size, color, and net). Results showed that the fruit swelling process (vertical and horizontal diameters) exhibited a slow-rapid-slow pattern, which could be well described with a logistic curve against growing degree days (GDD); fruit color changes based on the RGB values could be represented by quadratic relationship to cumulative GDD; the fruit net changes over growth progress could be partitioned into three phases according to the time interval. The first phase was from 1 to 30 days after pollination (DAP), in which the vertical stripe appeared at fruit middle part and the horizontal stripe at fruit petiole and hilum part as well; the second phase was from 30 to 40 DAP, the horizontal stripe occurred at fruit middle part and the net was formed; the third phase was the process started from 40 DAP, the netted breadth and thickness were gradually increased. The model was validated with the independent data from the experiment, and the mean RMSE (root mean square error) of fruit were 0.36 and 0.28 cm for vertical and horizontal diameters, 11.9 for fruit color, and 0.45 cm for stripe length and diameter at varied GDD, respectively.This work is beneficial to a reliable foundation for study the relationship between morphological formation and physiological change of the melon fruit internally and then realize the intelligent precision management to improve the yield and quality of greenhouse melon production.