Performance Analysis of Bird Counting Techniques Using Digital Photograph

Counting birds using digital photography is an attractive method. There are different methods of counting objects suitable to both offline and online real time situations. This research work has carried out a performance analysis of various methods utilized for bird census. This work is necessary to overcome some of the limitations of bird counting techniques that uses digital photography. It requires the study of existing techniques for object counting in a given digital photograph and to suggest some methods to overcome one or more limitations faced in traditional techniques. Counting techniques are most significant in various arenas of science and technology. The efficiency of manual counting drops with the increase in the number of objects. Counting bird using digital photography is most attractive, simple and less expensive. The manual counting techniques is the basic method. This method is not efficient as it human errors due to the involvement of human individuals and also not suitable for large flock of birds as human individual may miss some birds or recount the same bird. The proposed research work has explored new possibilities to overcome from the said problem by the application of the techniques such as contouring, thresholding, template matching, and counting objects. The results obtained from these studies are attractive and showed significant improvement over manual method of carrying out census of birds using digital photography.