Utilize Amazon SageMaker Geo Tools to Examine Rodent Infestations
Introduction:
Rodents such as rats and mice pose various health risks and are known to spread over 35 diseases. This article explores how to monitor and visualize rodent populations using Amazon SageMaker geospatial capabilities. By identifying regions of high rodent activity, local authorities and pest control organizations can effectively plan interventions and exterminate the rodents. The article demonstrates how to use satellite images and datasets to map rodent sightings, enrich location data with street addresses, and analyze the effects of rodent infestation on vegetation and bodies of water. Furthermore, it shows how to visualize the correlation between rodent sightings and cases of monkeypox. With SageMaker geospatial capabilities, data scientists and ML engineers can streamline the process of building, training, and deploying models using geospatial data.
Full Article: Utilize Amazon SageMaker Geo Tools to Examine Rodent Infestations
Monitoring and Visualizing Rodent Populations Using Amazon SageMaker Geospatial Capabilities
Introduction
Rodents, such as rats and mice, pose significant health risks and are known to spread over 35 diseases. Identifying regions with high rodent activity is crucial for local authorities and pest control organizations to effectively plan interventions and exterminate these pests. In this article, we explore how Amazon SageMaker geospatial capabilities can be used to monitor and visualize rodent populations. We will also examine the impact of rodent infestation on vegetation and bodies of water and analyze the correlation between rodent sightings and monkeypox cases.
Using Amazon SageMaker Studio Notebook
To begin, we utilize an Amazon SageMaker Studio notebook with a geospatial image. This image already includes SageMaker geospatial capabilities, making it easier to enrich data for geospatial analysis and machine learning (ML). We use satellite images from Sentinel-2 and the rodent activity and monkeypox datasets from NYC open data.
Accessing Geospatial Data
In this step, we extract the latitude and longitude of rodent sightings and inspections from the rodent activity dataset. We then enrich this location information with human-readable street addresses by running a reverse geocoding operation using Amazon Location Service. To do this, we create a vector enrichment job (VEJ) in the SageMaker Studio notebook. The VEJ converts the geographic coordinates to human-readable addresses and appends additional location-related columns to the dataset.
Visualizing Rodent Activity
Once the VEJ is complete, we export the output of the job to an Amazon S3 bucket. This output file contains the input data (longitude and latitude coordinates) along with the appended columns. With SageMaker geospatial capabilities, we can overlay the rodent sightings on a base map and visualize the data points along with their human-readable addresses. This visualization can aid in rodent inspection and treatment efforts.
Analyzing the Effects of Rodent Infestation
To analyze the effects of rodent infestation on vegetation and bodies of water, we classify each location as vegetation, water, or bare ground. SageMaker geospatial capabilities provide access to geospatial data such as Sentinel-2 and Landsat 8, saving significant time and effort in data collection and processing. We utilize a pre-trained Land Use Land Cover (LULC) segmentation model to identify the physical materials present at the earth’s surface. This analysis helps us understand how rodent populations affect vegetation and bodies of water.
Visualizing Monkeypox Cases and Correlating with Rodent Data
To visualize the correlation between monkeypox cases and rodent sightings, we add the monkeypox dataset along with the geoJSON file for New York City borough boundaries. Using the visualization tool powered by Foursquare in SageMaker Studio, we can add layers and charts to the map. We add the monkeypox data as a chart, showing the number of cases in each borough. Additionally, we overlay the borough boundaries as a polygon layer and the rodent activity as a heatmap layer. By comparing the visuals, we can identify patterns and correlations between rodent activity and monkeypox cases. For example, boroughs with a higher concentration of rodent sightings may also exhibit a higher number of monkeypox cases.
Conclusion
Amazon SageMaker geospatial capabilities provide essential tools for monitoring and visualizing rodent populations. By leveraging vector enrichment and Earth Observation Jobs (EOJs), along with built-in visualization tools, SageMaker makes it easier to handle large-scale geospatial datasets, perform model training and inference, and explore predictions on interactive maps. With this powerful platform, local authorities and pest control organizations can effectively plan interventions to manage rodent populations and minimize associated health risks.
Get Started with Amazon SageMaker Geospatial Capabilities
You can start utilizing Amazon SageMaker geospatial capabilities by visiting the official documentation and accessing example notebooks on SageMaker geospatial capabilities in the GitHub repository.
About the Authors
This article was written by Bunny Kaushik, a Solutions Architect at AWS, and Clarisse Vigal, a Sr. Technical Account Manager at AWS. Both authors are passionate about helping customers innovate on the AWS platform and accelerate their cloud adoption journey.
Summary: Utilize Amazon SageMaker Geo Tools to Examine Rodent Infestations
Rodents such as rats and mice pose significant health risks and are responsible for spreading numerous diseases. To effectively combat these pests, it is crucial to identify areas with high rodent activity. In this post, we explore how Amazon SageMaker’s geospatial capabilities can be used to monitor and visualize rodent populations. We also demonstrate how to analyze the impact of rodent infestations on vegetation and bodies of water. Additionally, we examine the correlation between rodent sightings and reports of monkeypox cases. With SageMaker, data scientists and ML engineers can easily access geospatial data sources, perform processing operations, apply ML models, and visualize the results.
