OpenCV.jl

OpenCV.jl provides Julia bindings to OpenCV, the popular computer-vision library — image and video processing, object detection, feature extraction, calibration, and much more — backed by the compiled library through OpenCV_jll.

Installation

] add OpenCV

This installs the package and the precompiled OpenCV binaries; no system OpenCV is required.

Versioning

This release wraps OpenCV 4.13.0, pinned in gen/OPENCV_VERSION and provided at runtime by OpenCV_jll. Query it from Julia with OpenCV.getVersionString() (or OpenCV.getBuildInformation() for the full build configuration):

julia> using OpenCV
julia> OpenCV.getVersionString()"4.13.0"

The package version itself tracks the wrapped OpenCV release. Within a given OpenCV major/minor line, breaking changes to the Julia API are signalled by bumping the package's minor version and called out in release notes — they do not require bumping the OpenCV major.

Quick start

using ImageCore, OpenCV, TestImages, MosaicViews

img_orig = testimage("Mandrill")

img_raw =  collect(rawview(channelview(img_orig)))

img_gray = OpenCV.cvtColor(img_raw, OpenCV.COLOR_RGB2GRAY)

img_blur = OpenCV.blur(img_gray, OpenCV.Size(Int32(9), Int32(9)))

img_back =  colorview(Gray, float.(img_blur)./255)[1,:,:]

mosaic(img_orig, img_back; nrow=1)

Output:

One thing to know up front: image layout

Images are Mats, 3-D arrays that share memory with OpenCV and use the axis order (channels, cols, rows) — so a grayscale H×W image has size == (1, W, H). This trips people up; see Core concepts for the full story.

Manual

Core concepts — naming (cv:: → OpenCV.), the Mat layout, dtypes, scalars, and Julia-array interop.
Reading and Writing of Images — imread/imwrite and the FileIO load/save interface, with flags and display.
API reference — the hand-written types and I/O helpers.
Wrapped functions — the complete generated list of wrapped OpenCV functions.

If you are new to OpenCV, the upstream OpenCV tutorials are the best starting point. Function and argument names mirror the C++/Python API, so most tutorials translate directly — the Julia-specific pieces are the Mat/Vec types and the array/image interop, all covered in Core concepts.