How to use this book
Mark Meagher
Each of the chapters in this book addresses a particular computational method and describes how to implement it using images collected in this project. Our aim has been to assume no previous knowledge of the methods presented here, although some familiarity with Python will be very helpful! If you’re not familiar with Python it would be helpful to follow a Python fundamentals training course – there are many great options online.
Our intention for this book and the accompanying resources to be used in teaching and learning is reflected in the structure and content of the resources. Taken as a whole the chapters provide an introduction to computational methods for working with large image collections, using images from the Understanding Animals project as an example of such a collection. Each chapter can also be used individually to introduce a particular technique or approach. These methods can be used across many types of images to address the multiple deficiencies of information that are typical of early design. Camera traps are ideal for understanding wildlife, and can also be used to capture other types of time-based information such as movements of people, fluctuations in environmental conditions such as water and wind, or documenting the weather. Each jupyter notebooks contains an introduction that outlines methods for adapting the techniques described to other types of image content.
We have found it practical to separate the technical details of implementing computational methods from a conceptual introduction. This pressbook contains the conceptual part, in which we describe why you might want to try these methods and how they can help gather information of relevance to designers. The technical information is located the jupyter notebooks which can be accessed on GitHub and in Google CoLab. The jupyter notebooks are interactive web-based documents that contain Python code, text descriptions of what is happening in the code, project images and images of outputs like graphs and tables. Each jupyter notebook is linked in the relevant pressbook chapter, and all the project jupyter notebooks can be browsed on GitHub.
Each Jupyter Notebook has a link at the top of the document to open the notebook in Google CoLab, an environment for interactively downloading the relevant data and running the Python code. CoLab is free for use but requires a Google account.
The data used in the Jupyter Notebooks is hosted with the University of Manitoba Dataverse. Each data source is linked directly to the Notebook in which it is used.
