Tscharke, M. and Banhazi, T. M. (2011) Review of methods to determine weight, size and composition of livestock from images. In: SEAg 2011: Diverse Challenges, Innovative Solutions, 29-30 Sep 2011, Gold Coast, Australia.
The currently available technologies which can determine the weight of livestock are reviewed and limitations of the weighing task by these means are defined. Comparisons between the different techniques highlight the superiority of the non-contact vision based method. Modelling techniques for weight estimation size and composition are reviewed along with image segmentation and recognition methods. Conclusions identify that more work is required on further minimising the error of group average daily weight estimates, individual animal weight estimates, group weight deviation and enhancing the design of the weighing apparatus to function in a commercially realistic environments. Future direction also centres on enhancing automation, minimising invasive environmental control, maximising precision and repeatability during dimension recovery and identifying and controlling the effect of any bias in weight estimation.
|Item Type:||Conference or Workshop Item (Commonwealth Reporting Category E) (Paper)|
|Additional Information:||Publisher does not formally support archiving.|
|Uncontrolled Keywords:||machine vision; allometry; computer vision; weight; image analysis; animals; livestock|
|Depositing User:||Dr Thomas Banhazi|
|Date Deposited:||05 Mar 2012 01:54|
|Last Modified:||28 Jun 2012 02:01|
Actions (login required)
|Archive Repository Staff Only|