Abstract:

The development of methods to identify microorganisms has always been helpful to scientists, but it has been particularly so in light of the several pandemics that have occurred in recent times. The need for novel and rapid detection approaches has been heightened by the ongoing increase in the global frequency of pandemics. Unexplored up to now, the present work suggests using Single-Shot Multibox Detector (SSD) for microbe detection. The 840 photos from 21 different microbial groups (40 images each) are part of the EMDS-6 Dataset, which is used by the suggested model. When it comes to identifying microorganisms, the selected model, SSD MobileNet V2 FPNLite 320x320, from the TensorFlow 2 model zoo, has shown results that are more than satisfactory in terms of performance, accuracy, and precision. The results demonstrate the adaptability and durability of this method based on SSDs. In addition to outlining the present project's future scope, the findings are also discussed.

Description:

.