Detecting egg laying

Worker egg-laying can be detected with the egg-laying_detector.jar program. This program analyzes observation hive images to estimate for each worker bee the probability that she has inserted her abdomen into a honeycomb cell, which is characteristic of honey bee egg-laying. Details of this procedure are described in this paper.

Detecting egg-laying

To detect egg-laying in the example image from the bCode detection page, we first need to rename and index the bCode detection file for this image as described in the trophallaxis detection tutorial. (If you have already done this, you can skip this step.) We can then use the indexed bCode detections to predict which worker bees are in the process of laying an egg

java -jar egg-laying_detector.jar input.file=2013-07-18-13-57-25-600.jpg filtered.data.file=bcode_detections.txt

This will result in a file named 2013-07-18-13-57-25-600.txt that contains one line for each bee in the format described on this page.

Hints

Note that the egg-laying detector expects filenames to encode the date and time when the image was captured. A valid file name has the format yyyy-MM-dd-HH-mm-ss-SSS, like in the example above. If you don't know what these letters mean, please look here.

To process more than one image, you can create a plain text file that lists one image on each line and then use this file as the input.file.

Since the egg-laying detector reports the probability of egg-laying, users can decide how high this probability needs to be for the detection to be considered "real". An occurrence threshold of 0.5 and a true positive cutoff of 0.55 result in the detection performance reported in our methods paper.

Limitations

The egg-laying detector was trained on images generated by the honey bee tracking rig described in this paper. If it is applied to images from other recording setups, its detection performance will likely differ from the one we reported.