September 01, 2011
Our paper: Automatic Analysis of Camera Image Data: An Example of Honey Bee (Apis cerana) Images from the Shanping Wireless Sensor Network. has been published in the Taiwan Journal of Science.
Under the international collaborative program between TFRI and PRIME of San Diego University USA in 2010, we extended an image analysis package and applied it to honey bee observations. In this article we describe the result of this collaboration. A tool suitable for routine measurements and counting tasks was developed to perform automatic process. We applied blob-detecting of computer vision technique to develop this package.
We then tested the tool by using different bee presence number images collected from Shanping wireless sensor network of TFRI. We compared the time consumed between the automatic process and manual process. Result shows that the low number of bee presence image (average bee number <30 individuals per image) analysis between automatic process and manual process are 9 and 315 mins.
A similar result shows that in the high number of bee presence image (average bee number >30 individuals per image) analysis between automatic process and manual process are 23 and 409 mins. Although the automatic process overestimates 2%-21% of bee counting, the tool shows significantly reducing the process time. We conclude that the program provides a convenient process to determine the target and thus facilitate the examination of large volume honey bee images from wireless sensor network of the field.