Weight estimation of Zandi lambs using digital image processing

This study was conducted to introduce a method based on using digital image processing to estimate the weight of new born lambs of Zandi sheep. Data was collected from 115 new born lambs reared in the Zandi sheep breeding centre of Khojir, Tehran. Primarily, all lambs were weighted, and at the same time, several photos were taken from the side view of each lamb using a digital camera from a fixed imaging distance. Then, the features related to the lateral area of lambs were extracted from digital images using image processing tools (IPT) of MATLAB software. Afterwards, a suitable artificial neural network was designed to estimate the lamb's weight based on the features was extracted from digital images. The neural network was trained with the precision of 96.94 % to estimate the lamb's weight. In the test phase, there was a correlation equal to 90.11 percent between the actual weight of lambs and weights estimated by the artificial neural network (p