Camera Trap Performance For Nocturnal Mammals by Kaggle | Technology and IT

About this Dataset

Camera Trap Performance For Nocturnal Mammals

Camera Trap Performance for Nocturnal Mammals

Insights into Flying Squirrel Populations in Urban Environments

About this dataset

This dataset captures data from motion detecting camera traps used to study flying squirrel (Glaucomys) populations in urban and natural environments. The traps use commercially available parts, cost-comparable to consumer camera traps, including a raspberry pi microcomputer, camera and 940 nm IR illuminator mounted on a tree-mounted wooden platforms. It contains records of the location, date of capture, number of captured events separated by type (flying squirrels, other mammals, insects and other wildlife types), trailcam model details and notes about each event. This data provides researchers with valuable insights into the habits of small nocturnal animals such as flying squirrels in both urbanized and natural habitats. It allows us to understand how animal behavior is affected by changes in their environment so that conservation efforts can be optimized for these species around the world

More Datasets

For more datasets, click here.

Featured Notebooks

🚨 Your notebook can be here! 🚨!

How to use the dataset

This dataset contains data collected from motion detecting camera traps used to study flying squirrels (Glaucomys) population in urban and natural environments. Using the data provided, you can analyze the performance of various types of cameras when capturing nocturnal wildlife. With this dataset, you will be able to answer questions such as: Which type of camera is most effective at capturing flying squirrels? Are there any correlations between location and captures? What are the differences in capture rates for different species?

In order to use this dataset, make sure to have a working knowledge of Python or another programming language. It may also be useful to familiarize yourself with types of camera traps in advance so that you can make best use of key elements such as ‘Cam’ and ‘Trailcam_model’ in your analyses.

To start off your analysis, begin by loading the data into a Pandas DataFrame for easy access and manipulation. You could then group by date across locations or different capture types (e.g FlyingSquirrel_captures) in order to get an idea of how certain locations perform better than others over time or relative trends around certain species captures versus another (such as FlyingSquirrel_captures vs Other_Mammal_captures). You could analyze other specific relationships using scatterplots or correlation matrices if needed.
To further dive into your questions regarding which type of camera works best for nocturnal mammal populations it would make sense explore the Trailcam_model column - e.g is there one model that consistently performs better than others? Note - utilizing other columns such as Total Captures might help solidify evidence towards which models are more efficient compared to others by giving us more evidence when voting on which ones are more capable than others.

Finally once all your research has been completed always remember to document key decisions made throughout your analysis such as assumptions made while creating visualizations or reasons why certain models captured more (or less ) activity than other models so that downstream readers can understand why certain conclusions were made if they occur while reviewing later on- this will help ensure repeatability down the line if changes must be made!

Research Ideas

Developing a better design for camera traps - Comparing the performance of different camera trap models and locations, as well as analyzing data like flying squirrel events and insect captures, can provide insights into how to better design a sensor that is optimized for small nocturnal mammals in urban areas.

Tracking Changing Environmental Conditions - Examining the total number of captures, insect captures, other mammal captures, and other events captured by the cameras over time can help identify trends in environmental conditions such as temperature changes or changes in vegetation due to deforestation or urbanization.

Developing AI/ML algorithms for automatic species recognition - Utilizing existing data from this study along with deep learning algorithms could potentially be used create automated species identification applications that utilize motion detector camera traps.

Acknowledgements

If you use this dataset in your research, please credit the original authors.
Data Source

License

License: CC0 1.0 Universal (CC0 1.0) - Public Domain Dedication
No Copyright - You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission. See Other Information.

Columns

File: Location.csv

Column name	Description
Site	The location of the camera trap. (String)
Cam	The type of camera trap used. (String)
lat	The latitude coordinate of the camera trap. (Float)
long	The longitude coordinate of the camera trap. (Float)

File: Data.csv

Column name	Description
Site	The location of the camera trap. (String)
Cam	The type of camera trap used. (String)
Date	Date of capture. (Date)
Flying_squirrel_detected_yes_or_no	Whether or not a flying squirrel was detected by the camera trap. (Boolean)
Total_captures	Total number of captures made with the camera trap. (Integer)
FlyingSquirrel_captures	Number of captures for flying squirrels. (Integer)
Other_Mammal_captures	Number of captures for other mammals. (Integer)
Insect_captures	Number of captures for insects. (Integer)
Other_captures	Number of captures for other animals. (Integer)
Trailcam_model	Model of trailcam used. (String)
trailcam_Flying_Squirrel_events	Number of flying squirrel events captured by the trailcam. (Integer)
trailcam_percent_of_Pi_cam_Flying_Squirrel	Percentage comparison with Pi cameras based on its Flying Squirrel signals. (Float)
trailcam_Other_Mammal_events	Number of other mammal events captured by the trailcam. (Integer)
trailcam_percent_of_Pi_cam_Other_Mammal	Percentage comparison with Pi cameras based on its other mammal signals. (Float)
trailcam_Other_events	Number of other events captured by the trailcam. (Integer)
notes_trailcams	Notes taken during the study. (String)

Acknowledgements

If you use this dataset in your research, please credit the original authors.
If you use this dataset in your research, please credit .

Tables

Data

@kaggle.thedevastator_camera_trap_performance_for_nocturnal_mammals.data

14.24 KB
48 rows
16 columns


CREATE TABLE data (
  "site" VARCHAR,
  "date" VARCHAR,
  "cam" BIGINT,
  "flying_squirrel_detected_yes_or_no" BIGINT,
  "total_captures" BIGINT,
  "flyingsquirrel_captures" BIGINT,
  "other_mammal_captures" BIGINT,
  "insect_captures" BIGINT,
  "other_captures" BIGINT,
  "trailcam_model" VARCHAR,
  "trailcam_flying_squirrel_events" DOUBLE,
  "trailcam_percent_of_pi_cam_flying_squirrel" DOUBLE,
  "trailcam_other_mammal_events" DOUBLE,
  "trailcam_percent_of_pi_cam_other_mammal" DOUBLE,
  "trailcam_other_events" DOUBLE,
  "notes_trailcams" VARCHAR
);

Location

@kaggle.thedevastator_camera_trap_performance_for_nocturnal_mammals.location

3.4 KB
16 rows
4 columns


CREATE TABLE location (
  "site" VARCHAR,
  "cam" BIGINT,
  "lat" VARCHAR,
  "long" VARCHAR
);