Image asymmetry measurement for the study of endangered Pygmy Bluetongue Lizard

There are applications for the measurement of body asymmetry as some studies have shown a correlation between asymmetry and fitness for some species. In our study of the endangered Pygmy Bluetongue Lizard, the asymmetry of its head is being investigated to see whether this has a correlation with its health and chance of survival in the wild. As there are restrictions on handling the endangered lizards, their digital photos must be taken in the field and therefore it is difficult to impose restrictions on the conditions under which the digital images are acquired. In this paper, we propose a novel automatic technique that is invariant to rotation, size, illumination and tilt, for the measurement of lizard symmetry based on its digital imagery and the resulting symmetry index is used to infer the lizard's asymmetry. The conventional manual methods being used by biologists for fluctuating asymmetry measurement have a number of disadvantages including human errors, and their methods of measurement are based on counting the number of scales and length measurement that do not often agree well with visual assessment. Our proposed image processing technique is non-invasive, robust in a way that will give a similar symmetric index for different images of the same lizard, and more importantly based on the actual image scale pattern of the lizards. Hence our proposed method will also give a better agreement with visual assessment.