Image Segmentation of Cattle Muzzle Using Region Merging Statistical Technic

Authors

  • Jullend Gatc Kalbis Institute of Technology and Business

DOI:

https://doi.org/10.21512/comtech.v6i4.2189

Keywords:

cattle muzzle, statistical region merging, segmentation, identification, digital image processing.

Abstract

Making an identification system that able to assist in obtaining, recording and organizing information is the first step in developing any kind of recording system. Nowadays, many recording systems were developed with artificial markers although it has been proved that it has many limitations. Biometrics use of animals provides a solution to these restrictions. On a cattle, biometric features contained in the cattle muzzle that can be used as a pattern recognition sample. Pattern recognition methods can be used for the development of cattle identification system utilizing biometric found on the cattle muzzle using digital image processing techniques. In this study, we proposed cattle muzzle identification method using segmentation Statistical Region Merging (SRM). This method aims to identify specific patterns found on the cattle muzzle by separating the object pattern (foreground) from unnecessary information (background) This method is able to identified individual cattle based on the pattern of it muzzle. Based on our evaluation, this method can provide good performance results. This method good performance can be seen from the precision and recall : 87% and the value of ROC : 0.976. Hopefully this research can be used to help identify cattle accurately on the recording process.

Dimensions

Plum Analytics

References

A. Novianto, A. M. Arymurthy. (2012a). Beef Cattle Identification Based On Muzzle Pattern Using a Matching Refinement Technique in SIFT Method. ICACSIS.

A. Novianto & A. M. Arymurthy. (2012b) Automatic Cattle Identification based on Muzzle Photo Using Speed-Up Robust Features Approach. 3rd European Conference of Computer Science ECCS.

Boltz, S. (2009). Image Segmentation Using statistical Region Merging. IEEE Trans. Pattern Anal. Mach. Intell, 26: 1452-1458.

B. Zhang, et. al. (2010). Retinal Vessel Extraction by Matched Filter with First-Order Derivative of Gaussian. Computers in Biology and Medicine, 40: 438-445.

D. P. Berry., et. al. (2005) Accuracy of Predicting Milk Yield From Alternative Milk Recording Schemes. Animal Science, 80(01): 53-60.

F. Lang., et. al. (2014). Polarimetric SAR Image Segmentation Using Statistical Region Merging. Geoscience and Remote Sensing Letter, 11(2): 509-513.

Gonzalez, R. C., et. al. (2008) Digital Image Processing. 3rd ed. USA: Prentice Hall. 2008.

Gualtieri, C. T., Ondrusek, M. G. and Finley, C. (1985). Attention deficit disorders in adults. Clinical Neuropharmacology, 8: 343-356.

H. Leclerc., et.al. (2004). Milk Recording: a Comparison Of The T, Z and Standard Method (Z = Milk Yield Recorded on 2 Milkings and The Contents on One Alternate Milking). Performance Recording of Animals: State of the Art: 198-205.

J. Gatc., et. al. (2012). Plasmodium Parasite Detection on Red Blood Cell Images for the Diagnosis of Malaria Using Double Thresholding. International Conference on Advance computer Science and Information System: 731- 736.

Nock, R., Nielsen. F. (2014) Statistical Region Merging. IEEE Trans. Pattern Anal. Mach. Intell, 26: 1452-1458.

Rahayu, A., et al. (2013). Genetic Evaluation of Dairy Cattle Using Actual Milk Yield Records and Centering Date Method (CDM). Jurnal Ilmiah Peternakan, 1(1): 236-243.

Zhai, L. & Hu, Q. (2011). The Research of Double-biometric Identification Technology Based on Finger Geometry and Palm Print. 2nd International Conference Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC). IEEE xplore: 3530-3533.

Downloads

Published

2015-12-01

Issue

Section

Articles
Abstract 471  .
PDF downloaded 368  .