Application of data fusion technique to fruiter profile modeling spray

It is necessary to exactly detect the shape parameters of fruiter for fruiter profile modeling spray. The shape parameters of fruiter was detected by image processing, and based on the leaf density the influence of imaging distance on detecting the leaf density in the processing of image was studied. For eliminating above influence, the data fusion between leaf density and imaging distance was processed by using BP artificial neural network and data fusion model. The results indicated that the minimal relative error between data fusion value and theoretical value is 0.19%, the maximal relative error is 5.31%. The data fusion value can be used for spraying control in fruiter profile spray system.