Computer Support for Tracking, Tracing and Quality ... - PanAFITA

5 downloads 6529 Views 166KB Size Report
computer vision techniques for beef maturity assessment (Figure 1). Çiçek and Şenkul (2006) evaluated the geographical information systems and it's usage ...
USE OF IMAGE ANALYSIS IN ANIMAL SCIENCE Hasan Önder1, Arzu Arı2, Sezen Ocak3, Samet Eker4 and Hacer Tüfekçi5 1

Ondokuz Mayis University, Faculty of Agriculture, Dep. of Animal Science, Samsun-Turkey,[email protected]

2

Ondokuz Mayis University, Faculty of Agriculture, Dep. of Animal Science, Samsun-Turkey , [email protected] 3

41 5

Cukurova University, Faculty of Agriculture, Dep. of Animal Science, Adana-Turkey, [email protected]

Ondokuz Mayis University, Faculty of Agriculture, Dep. of Animal Science, Samsun-Turkey, [email protected]

Ondokuz Mayis University, Faculty of Agriculture, Dep. of Animal Science, Samsun-Turkey,[email protected]

Abstract Image analysis and belonging biometric techniques have been rapidly increased in last decade. In this review, use of image analysis in animal science was described. Methodology and modeling of the image analysis of recent methods was introduced. Biometric technologies for this aspect have been introduced. Some of the use areas of image analysis such as determination of egg and meat quality in animal science was summarized.

Keywords: image analysis, animal science

Introduction Over the past few decades, biometric vendors have worked on increasing their levels of accuracy by tweaking both the algorithms doing the biometric mapping and matching, and the technology taking the readings. When considering the use of biometrics it can be difficult to determine which option is the best. This is largely due to the fact that there is no perfect biometric solution. Each one (such as voice recognition and face recognition) has its benefits and disadvantages (Moore, 2007). For digital image analysis, image quality is the most important factor. Depending on the number of traits, sensors, and feature sets used, a variety of scenarios are possible in a multimodal biometric system. To take detailed information please interest with Ross and Jain, 2004. Used algorithms can be image or movie based. To analyze the digital image some models such as principal component analysis linear discriminant analysis, elastic bunch graph matching and hidden Markov model can be used (Önder, 2007). Machine vision systems have been used for agricultural and food applications, particularly in grading and inspection. These systems are useful for performing tedious or repetitive work. Image-processing techniques make it possible to identify and classify agricultural commodity not only in the spatial domain, but also in the spectral domain (Hatem et al., 2003).

Use in Animal Science Digital image analysis can be used in animal science with many aspects. Some approaches were given below; Hatem at al. (2003) used image processing to determine the beef maturity and they showed the potential of computer vision techniques for beef maturity assessment (Figure 1). Çiçek and Şenkul (2006) evaluated the geographical information systems and it’s usage possibilities in livestock sector. Ozder et al. (2007) investigated the Utilizing possibility of Image Process Technology by using methods of Digital Image Process on the estimation and Digital Image Analysis on evaluation of live weight and various body measurements of slaughter cattle. The results of this study showed the presence of positive and high correlations between live weight and body measurements and also facilities of using in the practice of these calculated body measurements by this method. Işık and Güler (2009) used image processing to determine teat apex deformation on different vacuum levels (Figure 2).

Vol(1)

Journal of Information Technology in Agriculture

1

Figure 1. An illustration of the ossification feature curve derivation for A-maturity sample (a) and E-maturity sample (b) (Hatem at al., 2003).

Figure 1. an example image of teat apex deformation (Işık and Güler, 2009). Buche and Mauron (1997) applied image analysis to quantitative characterization of muscle fibre. They declared that to quantitative characterization of muscle fiber can be successfully determined and origin of the meet (races) can be recognized. Lu and Tan (2004) used image based predictors to predict beef lean yield. They identified the image processing techniques for meet quality. İnce and Ayhan (2008) reviewed the carcass quality prediction in sheep by image analysis. Pipek et al. (2004) used of video image analysis for fat content estimation of cattle carcasses (Figure 3). They argued that the fat content determined by Soxhlet extraction method appears to be in good accordance with the results obtained by video image analysis. Meat image shooting and subsequent discrimination of the adipose tissue from other tissues (connective, bones) and a voiding of distractive gloss effects are limiting factors.

Figure 1. Thresholding of adipose tissue (Pipek et al. 2004) Vol(1)

Journal of Information Technology in Agriculture

2

Smékal et al. (2005) used image analysis to evaluate the beef carcasses. They found that the areas of muscle and adipose tissues were thresholded on several ways (the whole loin area, the areas of individual muscles and those of different loin sections) and the correlations between these areas were searched. The best correlations were found between large areas. Video image analysis proved to be a suitable method for the evaluation of selected carcass parts. Polák et al. (2008) predicted of intramuscular fat in live bulls using real-time ultrasound and image analysis. They declared that the obtained results suggest that prediction of intramuscular fat in meat is possible using the technique of ultrasound and computer image analyses of live animals. Aktan (2004a) used image analysis to determine some exterior and interior quality traits of quail eggs and phenotypic correlations. Wang et al. (2009) used digital image technology to detect the egg freshness. They argued that The egg freshness detection based on image characteristic of yolk and air room was efficient and feasible. Davies et al. (2006) focused to investigate whether or not there were correlations between ultrasound image attributes of the corpus luteum (CL) and changing progesterone concentrations over time, in prolific and non-prolific ewes. They exposed that there was no significant correlation between progesterone concentrations and spot pixel heterogeneity for either Western White Face ewes or Finn ewes. Aktan (2004b) determined the some carcass characteristics by digital image analysis in broiler chickens. They found expressive that the color variations between the bruises that were visually classified as mostly red, mostly purple, and mostly green in color for grey shade values of RGB components, and therefore these differences could be digitally and recognizably described. Rashidi et al. (2008) used image processing and spheroid approximation to determine egg volume. They declared that both methods yield similar results on egg volume determination. Tillett et al. (1997) studied on tracking animal movements. As well as position and rotation, more subtle motion such as bending and head nodding can be modelled. They claim that this type of model based tracking could be used to characterize animal behaviour over time. Dugnol et al. (2007) investigate the use of image processing techniques based on partial differential equations applied to the image produced by time–frequency representations of one-dimensional signals, they applied spectrogram image processing technique both for synthetic signals and for wolves chorus field recorded signals, which was the original motivation of their work. Serdaroğlu and Purma (2006) used image analysis to determine the quality of seafood. They examined the physical, mechanical, rheological methods like electrical measurements, optical measurements, texture and image analysis and measurements of volatile compounds Severa et al. (2010) used image analysis to evaluate of shape variability of stallion sperm heads. They declared that the phenomenon of stallion sperm head shape variability can be examined by image processing. Düzgün and Or (2009) used thermal camera and image processing on veterinary. They aimed to determine the using of thermal camera in medical and veterinarian showing the advantages of thermography.

Conclusions It is easily commented that image analysis is a rapidly increasing tool for animal sciences. Especially it is used for meat quality, but it can be used various aims. Image analysis can be used to determine the mastitis and oestrus determination.

References Aktan, S., 2004a. Determining some exterior and interior quality traits of quail eggs and phenotypic correlations by digital image analysis. Hayvansal Üretim 45(1): 7-1.3. Aktan, S., 2004b. Determining some carcass characteristics by digital image analysis in broiler chickens. Hayvansal Üretim 45(1):14-18. Buche, P and D. Mauron, 1997. Quantitative characterization of muscle fiber by image analysis. Computers and Electroncis in Agriculture 16: 189-217. Çiçek, H. and Ç. Şenkul, 2006. Geographical information systems and it’s usage possibilities in livestock sector. Veteriner Hekimler Derneği Dergisi 77(4): 32 – 38. Davies, K. L., P. M. Bartlewski, L. A. Pierson and N. C. Rawlings, 2006. Computer assisted image analyses of corpora lutea in relation to peripheral concentrations of progesterone: A comparison between breeds of sheep with different ovulation rates. Animal Reproduction Science 96: 165–175. Düzgün, D.and M. E. Or, 2009. Application of thermal camera on the medicine and veterinary. Tubav Bilim Dergisi 2(4): 468 – 475.

Vol(1)

Journal of Information Technology in Agriculture

3

Dugnol, B., C. Fernandez and G. Galiano, 2007. Wolf population counting by spectrogram image processing. Applied Mathematics and Computation 186: 820–830. Hatem, I., J. Tan, and D. E. Gerrard, 2003. Determination of animal skeletal maturity by image processing. Meat Science 65: 999–1004. İnce, D. and V. Ayhan, 2008. Carcass quality prediction methods in sheep. Hayvansal Üretim 49(1): 57-61. Işık, E. and T. Güler, 2009. Determination of teat apex deformation on different vacuum level by image analyses method. Journal of Agricultural Faculty of Uludag University 23(1): 33-41. Lu, W. and J. Tan, 2004. Analysis of image-based measurements and USDA characteristics as predictors of beef lean yield. Meat Science 66: 483–491. Moore, A.M., 2007. Biometric technologies – an introduction. biometric technology today, 15(1): 6-7. Önder, H., 2007. Biyometrik teknolojiler. Popüler Bilim, 14(158), 32-36. Ozder, M., O. K. Dogaroglu, Y. T. Tuna, and A. R. Onal, 2007. Utilizing possibility of image process technology on the estimation of live weight and various body measurements of slaughtery cattles. 3rd Joint Meeting of the Network of Universities and Research Institutions of Animal Science of the South Eastern European Countries, Thessaloniki 10-12 February 2007. Pipek, P., J. Jelenikova, and L. Sarnovsky, 2004. The use of video image analysis for fat content estimation. Czech J. Anim. Sci. 49(3): 115–120. Polák, P., J. A. Mendizabal, N. E. Blanco Roa, E. Krupa, J. Huba, D. Peškovičová, and M. Oravcová, 2008. Prediction of intramuscular fat in live bulls using real-time ultrasound and image analysis. Journal of Animal and Feed Sciences 17: 30–40. Rashidi, M., M. Malekiyan, and M. Gholami, 2008. Egg volume determination by spheroid approximation and image analysis. Worl Applied Science Journal 3(4): 590 – 596. Ross, A. and K. Jain, 1994. Multimodal biometrics: An overview. 12. European Signal Processing Conference, 1221 – 1224, Viyana, Avusturya. September. Serdaroğlu, M. and Ç. Purma, 2006. Rapid techniques used to determine the quality of seafood. E.U. Journal of Fisheries & Aquatic Sciences 23 (1/3): 495 – 496. Severa, L, L. Machal, L. Svabova, and O. Mamica, 2010. Evaluation of shape variability of stallion sperm Heads by means of image analysis and fourier descriptors. Animal Reproduction Science 119: 50–55. Smékal O., P. Pipek, M. Miyahara and J. Jeleníková, 2005. Use of video image analysis for the evaluation of beef carcasses. Czech J. Food Sci., 23: 240–245. Tillett, R. D., C. M. Onyango and J. A. Marchant, 1997. Using model-based image processing to track animal movements. Computers and Electronics in Agriculture 17: 249 – 261. Wang, Q., X. Deng, Y. Ren, Y. Ding, L. Xiong, Z. Ping, Y. Wen and S. Wang, 2009. Egg freshness detection based on digital image technology. Scientific Research and Essay 4(10): 1073-1079.

Vol(1)

Journal of Information Technology in Agriculture

4

Suggest Documents