For greater efficiency in detection, in order to prevent late actions against pests, Amberjack has a smart trap as a solution. Applying artificial intelligence algorithms to identify insects attracted by the pheromone assigned to the trap.
However, if the needed species is not within the database , the preparations of the algorithm training is started. Consisting of the daily transfer of an image of the adhesive surface, in the event of the presence of objects or insects.
This image is received by a biologist and analyzed in order to alert the presence of pests, as well as the identification of species. This is done in order to increment a dataset of the species, with the ultimate goal of training a model based on this data. After the completion of the model training, the automatic detection of insects will be evaluated by the biologist to validate the confidence level of the object detector and whether the detected species are accurate.
The traps are directly connected to the dataloggers, and the connection can be made via USB, RJ45, or wireless.
The entrance of the trap has a wide opening with the purpose of encompassing a wider range of insects, such as the Phthorimaea operculella, commonly known as the potato tuber moth. Adult moths are attracted to the yellow and green colors and will fly towards it, thus becoming attached to the adhesive surface. Traps were designed with these colors in mind to aid their effectiveness in tracking and controlling pests in fields or storage facilities.
– Amberjack Datalogger, specified for AI algorithms.
– 12, 16, or 64MP depending on the requirements and species.
– USB or Ethernet connection.
– Contact us for a complete assessment.