A REGRESSION MODEL FOR NONDESTRUCTIVE FRUIT VOLUME ESTIMATION IN KARONDA (CARISSA CARANDUS ARANDUS L.)

Abstract

A model relating to karonda (Carissa carandus.L.) fruit length and diameter to its volume was derived using the regression analysis. Two year investigation was carried out during 2014 and 2015 under open field conditions to test a whether a model could developed to estimate fruit volume using linear measurements. Regression analysis having FV versus L and D disclosed ten models viz., FV= -4.52 + 0.5(L+D) 2, FV= -18.14 + 5.5(L+D), FV= - 48.19 + 37.22D0.5, FV= -41.56 + 32.09L0.5, FV=2.07+ 0.15L2D2, FV= -5.37+ 2.44D2 , FV= -3.44+1.88L2, FV= -4.55+2.18LD, FV=-19.62+12.02D and FV=-16.08+9.99L that could be used for estimating the volume of individual karonda fruits. Among these models a regression model having LD as the independent variable (FV = -4.55 +2.18 *LD) provided most accurate (R2=0.98), MSE=0.75) estimate of karonda FV. Validation of the model having LD of fruits from other genotype measured in the 2015 experiment showed that the correlation between calculated and measured area was very high. Therefore, this model can estimate accurately the FV of karonda in many experimental comparisons without the use of any pricey instruments.

Allometric model fruit