Computer-assisted image analysis for the comparison of sampling methods for assessing wool properties in Arkharmerino × Ghezel crossbreed sheep

Abstract

Usually investigations on wool properties are carried out using classical instruments like a Projection Microscope at wool laboratories in Iran. It is readily acknowledged that the Projection Microscope is of limited usefulness. Firstly, it is a very slow and labour intensive technique. Secondly, it is very imprecise when a single operator conducts measurements within one laboratory. Image analysis presents a quick, reliable, unbiased technique for determining wool properties (Rose and Debra, 2001). The application of computer-assisted image analysis to the measurement of wool properties potentially enables a significant reduction in the counting time for each sample. Recently in Iran, coarse-mixed wools have been valued and used in similar areas as finer wools. Coarse-mixed wools can be blended with artificia1 fibres to produce high valued clothes. They are also used widely in carpet production. This is why, in Iran, carpet wool breeds of increasing fleece weight is a main selection target, with improvement of fineness and uniformity as the next most important. Typically small samples from the whole fleece are taken using a scissors. The samples can be taken from six regions, i.e. neck, back, hip, shoulder, flank and rump, however usually they are taken from three regions, namely the shoulder, last rib and rump, in breeds with coarse-mixed wools. These sampling methods cause loss of time and require more labour. In this research, two methods of sampling the Arkharmerino × Ghezel fleece were compared. Sampling from only over the right mid-side, as occurs in fine wool breeds was compared with taking samples from three regions (shoulder, last rib and rump).