Table 1.
Terms used by different groups practising using advanced and new technology to develop CT and references to find more information about each term.
term | definition | reference |
---|---|---|
application programming interface (API) | a set of rules that allow applications and programs to communicate with each other | Boateng et al. [30] |
back-end interface | the server and work behind-the-scenes to allow the user interface | Smith [31] |
image classification | the computer vision process of predicting a class of one object to an image | Krizhevsky et al. [32] |
colonization of conservation | the historical legacy that is conservation is performed by those that colonized the areas where the conservation is performed | Loss et al. [33] |
conservation technology | an interdisciplinary field that works to design technology to help prevent the sixth mass extinction | Berger-Tal & Lahoz-Monfort [1] |
conservation tools (CT) | devices that are made and developed to be applied to the conservation of wildlife | this study |
context of use | a design thinking that takes the exact use of the device as the primary design component | Jacobson [34] |
ex situ conservation | conservation of a species outside its original place (e.g. in a zoo) | Braverman [35] |
fine-tuning | the computer vision process of taking a model that has been trained on one task and tuning it to make it perform a different, similar task | this study |
front-end interface | the interface that the user sees, sometimes described as the user interface | Smith [31] |
frugal science | the concept of creating scientific tools that are the most accessible possible in the form of cost and functionality | Byagathvalli et al. [36] |
graphical user interface (GUI) | a digital interface where a user can interact with various components such as buttons or text boxes | Edler et al. [37] |
human-centered design | a design thinking that takes the context-of-use of the exact devices as the primary design component | Jacobson [34] |
human–wildlife centred design (HWCD) | using the human–wildlife interaction in the design process similar to that of human-centred design | this study |
indigenous design | a design thinking that is designed by the indigenous population that is most familiar with the conservation and ecological initiatives | Nawrotski & Kadatska [38] |
in situ conservation | conservation of a species at the original place (e.g. in the wild) | Braverman [35] |
object detection | the computer vision process of detecting instances of semantic objects of a particular class or set of classes | Lin et al. [39] |
object tracking | the computer vision task of taking a set of initial object detections, creating a unique identifier for each detection, and tracking each object over a series of time | Yilmaz [40] |
object re-identification (Re-ID) | takes object detection one step further by matching a given object in a new environment to the same object in a different environment | Stewart et al. [41] |
open-source solutions | solutions that are open access and solutions that are fully accessible by the public to re-create, re-design and re-invite | Lerner & Tirole [42] |
opportunistic technology | devices that are built for a particular industry, such as camera traps designed for hunters, but used in a different purpose, such as biologists using camera traps for ecological surveys | Berger-tal & Lahoz-Monfort [1] |
silver-bullet solutions | a one-size-fits-all solution that can address and solve any issue | Shaw [43] |
self-supervised learning | a machine learning subset in which a model trains itself to learn part of the input from another part of data, often leveraging the underlying structure of the data | Hendrycks et al. [44] |
supervised learning | a machine learning subset of problems where the available data has labelled examples | Russell & Norvig [45] |
transfer learning | a machine learning method that uses a pre-trained model as a starting point for a model in a new task (i.e. it has already learned how to ‘see’ one set of things and will be trained again to get better focus on another set of things) | Zhuang et al. [46] |
unsupervised learning | a machine learning subset of problems that analyses and clusters unlabelled data | Schmarje et al. [47] |
wildlife | collective term referring to non-domesticated species of animals, plants and microbes, though sometimes restricted to just mean animals (particularly mammals and birds) | Usher [48] |