It is worth noting that determining the shape parameters of a fruit growth period, monitoring fruit development, and solving the problem of fruit characteristic size extraction for quality grading are essential in agricultural fields. However, traditional measurement methods are prone to damage fruit surface morphology, and the noncontact measurement method based on two-dimensional image feature processing has limitations in measuring various morphological parameters (e.g., size and volume). To solve this problem, this paper proposes an adaptive co-opposite-direction slicing noncontact measurement method for the orange shape parameters. First, we use an active structured light three-dimensional camera to obtain the orange point cloud. Consequently, we calculate the height and diameter of the orange point cloud using the search bounding box method. Furthermore, to improve the slice utilization rate, we introduce adaptive conditions to examine the relationship between the change rate of adjacent slices' vertical area and the threshold value to automatically update the slices. After that, the polygon area is updated again to calculate the slice volume. Finally, the complete volume of the orange is calculated using the summation method, and the final calculated value is determined by the mean of the volume obtained through co-opposite-direction calculation. Based on the verification of the orange simulation model and two groups of real orange data sets, the results of regression analysis between the calculated and real values show that the coefficient of determination of each index is greater than 0.95, and the average time is no more than 8.354 s. Compared with other methods, the measurement errors of height, diameter, and volume of the orange simulation model are reduced by 3.5%, 0.9%, 0.7%, and 3.6%.