Package: animl 1.1.0

Kyra Swanson

animl: A Collection of ML Tools for Conservation Research

Functions required to classify subjects within camera trap field data. The package can handle both images and videos. The authors recommend a two-step approach using Microsoft's 'MegaDector' model and then a second model trained on the classes of interest.

Authors:Kyra Swanson [aut, cre], Mathias Tobler [aut]

animl_1.1.0.tar.gz
animl_1.1.0.tar.gz(r-4.5-noble)animl_1.1.0.tar.gz(r-4.4-noble)
animl.pdf |animl.html✨
animl/json (API)

# Install 'animl' in R:
install.packages('animl', repos = c('https://cran.r-universe.dev', 'https://cloud.r-project.org'))

Peer review:

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

34 exports 0.00 score 57 dependencies 9 scripts 250 downloads

Last updated 1 years agofrom:a6d6204cf0. Checks:OK: 2. Indexed: yes.

TargetResultDate
Doc / VignettesOKAug 27 2024
R-4.5-linuxOKAug 27 2024

Exports:%>%animlapplyPredictionsbestGuessbuildFileManifestcheckFileconvertCoordinatescropImageGeneratorcropImageTrainGeneratordetectObjectdetectObjectBatchextractBoxesextractBoxesFromFlatextractBoxesFromMDgetAnimalsgetEmptyImageGeneratorImageGeneratorSizeimagesFromVideosloadDataloadMDModelparseMDplotBoxespredictSpeciesprocessYOLO5resizePadsaveDatasequenceClassificationsetupDirectorysymlinkMDsymlinkSpeciessymUnlinktestMDupdateResults

Dependencies:avbackportsbase64encbmpcliconfigcpp11curldigestdownloaderdplyrexifrfansigenericsgluehereigraphimagerjpegjsonlitekeraslatticelifecyclemagrittrMatrixpbapplypillarpkgconfigplyrpngprocessxpspurrrR6rappdirsRcppRcppTOMLreadbitmapreticulaterlangrprojrootrstudioapistringistringrtensorflowtfautographtfdatasetstfrunstibbletidyselecttiffutf8vctrswhiskerwithryamlzeallot

Readme and manuals

Help Manual

Help pageTopics
Titleaniml
Apply Classifier Predictions and Merge DataFramesapplyPredictions
Select Best Classification From Multiple FramesbestGuess
Extract exif Data and Create File ManifestbuildFileManifest
Check for files existence and prompt user if they want to loadcheckFile
Convert bbox from Relative to Absolute CoordinatesconvertCoordinates
Tensorflow data generator that crops images to bounding box.cropImageGenerator
Tensorflow data generator for training that crops images to bounding box.cropImageTrainGenerator
Run MD on a Single ImagedetectObject
Run MegaDetector on a batch of imagesdetectObjectBatch
Extract bounding boxes and save as new image from a batch of imagesextractBoxes
Extract crops from a single image represented by a processed dataframeextractBoxesFromFlat
Extract bounding boxes for a single image and save as new imagesextractBoxesFromMD
Return a dataframe of only MD animalsgetAnimals
Return MD empty, vehicle and human images in a dataframegetEmpty
Perform image augmentation through random color adjustments on an image/label pair.imageAugmentationColor
Perform random geometric transformations on an image.imageAugmentationGeometry
Tensorflow data generator that resizes images.ImageGenerator
Tensorflow data generator that resizes images and returns original image size.ImageGeneratorSize
Load image and return a tensor with an image and a corresponding label.imageLabel
Load image, crop and return a tensor with an image and a corresponding label.imageLabelCrop
Extract frames from video for classificationimagesFromVideos
Load .csv or .Rdata fileloadData
Load an image and return the full size image as an image tensor.loadImage
Load and resize an image and return an image tensor.loadImageResize
Load, resize and crop an image and return an image tensor.loadImageResizeCrop
Load and resize an image and return an image tensor as well as a tensor with the original image size.loadImageResizeSize
Load MegaDetector model file from directory or fileloadMDModel
parse MD results into a simple dataframeparseMD
converte the JSON file produced bye the Python version of MegaDetector into the format produced by detectObjectBatchparseMDjson
Plot bounding boxes on image from md resultsplotBoxes
Classifies Crops Using Specified ModelspredictSpecies
Process YOLO5 output and convert to MD formatprocessYOLO5
Resize an image with paddingresizePad
Save Data to Given FilesaveData
Leverage sequences to classify imagessequenceClassification
Set Working Directory and Save File Global VariablessetupDirectory
Create SymLink Directories and Sort Classified Images Based on MD ResultssymlinkMD
Create SymLink Directories and Sort Classified ImagessymlinkSpecies
Remove SymlinkssymUnlink
Select a Random Image and Run Through MegaDetectortestMD
TitleupdateResults