The GoCV package supports the latest releases of Go and OpenCV (v4.9.0) on Linux, macOS, and Windows. We intend to make the Go language a "first-class" client compatible with the latest developments in the OpenCV ecosystem.
GoCV supports [CUDA](https://en.wikipedia.org/wiki/CUDA) for hardware acceleration using Nvidia GPUs. Check out the [CUDA README](./cuda/README.md) for more info on how to use GoCV with OpenCV/CUDA.
GoCV also supports [Intel OpenVINO](https://software.intel.com/en-us/openvino-toolkit). Check out the [OpenVINO README](./openvino/README.md) for more info on how to use GoCV with the Intel OpenVINO toolkit.
## How to use
### Hello, video
This example opens a video capture device using device "0", reads frames, and shows the video in a GUI window:
This is a more complete example that opens a video capture device using device "0". It also uses the CascadeClassifier class to load an external data file containing the classifier data. The program grabs each frame from the video, then uses the classifier to detect faces. If any faces are found, it draws a green rectangle around each one, then displays the video in an output window:
```go
package main
import (
"fmt"
"image/color"
"gocv.io/x/gocv"
)
func main() {
// set to use a video capture device 0
deviceID := 0
// open webcam
webcam, err := gocv.OpenVideoCapture(deviceID)
if err != nil {
fmt.Println(err)
return
}
defer webcam.Close()
// open display window
window := gocv.NewWindow("Face Detect")
defer window.Close()
// prepare image matrix
img := gocv.NewMat()
defer img.Close()
// color for the rect when faces detected
blue := color.RGBA{0, 0, 255, 0}
// load classifier to recognize faces
classifier := gocv.NewCascadeClassifier()
defer classifier.Close()
if !classifier.Load("data/haarcascade_frontalface_default.xml") {
fmt.Printf("start reading camera device: %v\n", deviceID)
for {
if ok := webcam.Read(&img); !ok {
fmt.Printf("cannot read device %v\n", deviceID)
return
}
if img.Empty() {
continue
}
// detect faces
rects := classifier.DetectMultiScale(img)
fmt.Printf("found %d faces\n", len(rects))
// draw a rectangle around each face on the original image
for _, r := range rects {
gocv.Rectangle(&img, r, blue, 3)
}
// show the image in the window, and wait 1 millisecond
window.IMShow(img)
window.WaitKey(1)
}
}
```
### More examples
There are examples in the [cmd directory](./cmd) of this repo in the form of various useful command line utilities, such as [capturing an image file](./cmd/saveimage), [streaming mjpeg video](./cmd/mjpeg-streamer), [counting objects that cross a line](./cmd/counter), and [using OpenCV with Tensorflow for object classification](./cmd/tf-classifier).
You can use `make` to install OpenCV 4.9.0 with the handy `Makefile` included with this repo. If you already have installed OpenCV, you do not need to do so again. The installation performed by the `Makefile` is minimal, so it may remove OpenCV options such as Python or Java wrappers if you have already installed OpenCV some other way.
The following commands should do everything to download and install OpenCV 4.9.0 with CUDA and OpenVINO on Linux. Make sure to change `$HOME/folder/with/your/src/` to the directory you used to clone GoCV:
If you have already done the "Quick Install" as described above, you do not need to run any further commands. For the curious, or for custom installations, here are the details for each of the steps that are performed when you run `make install`.
By default, pkg-config is used to determine the correct flags for compiling and linking OpenCV. This behavior can be disabled by supplying `-tags customenv` when building/running your application. When building with this tag you will need to supply the CGO environment variables yourself.
Please note that you will need to run these 2 lines of code one time in your current session in order to build or run the code, in order to setup the needed ENV variables. Once you have done so, you can execute code that uses GoCV with your custom environment like this:
The project now provides `Dockerfile` which lets you build [GoCV](https://gocv.io/) Docker image which you can then use to build and run `GoCV` applications in Docker containers. The `Makefile` contains `docker` target which lets you build Docker image with a single command:
By default Docker image built by running the command above ships [Go](https://golang.org/) version `1.20.2`, but if you would like to build an image which uses different version of `Go` you can override the default value when running the target command:
Sometimes your `GoCV` programs create graphical interfaces like windows eg. when you use `gocv.Window` type when you display an image or video stream. Running the programs which create graphical interfaces in Docker container on macOS is unfortunately a bit elaborate, but not impossible. First you need to satisfy the following prerequisites:
Note, you will have to log out and log back in to your machine once you have installed `xquartz`. This is so the X window system is reloaded.
Once you have installed all the prerequisites you need to allow connections from network clients to `xquartz`. Here is how you do that. First run the following command to open `xquart` so you can configure it:
Next, you need to create a TCP proxy using `socat` which will stream [X Window](https://en.wikipedia.org/wiki/X_Window_System) data into `xquart`. Before you start the proxy you need to make sure that there is no process listening in port `6000`. The following command should **not** return any results:
You are now finally ready to run your `GoCV` GUI programs in Docker containers. In order to make everything work you must set `DISPLAY` environment variables as shown in a sample command below:
```shell
docker run -it --rm -e DISPLAY=docker.for.mac.host.internal:0 your-gocv-app
```
**Note, since Docker for MacOS does not provide any video device support, you won't be able run GoCV apps which require camera.**
### Alpine 3.7 Docker image
There is a Docker image with Alpine 3.7 that has been created by project contributor [@denismakogon](https://github.com/denismakogon). You can find it located at [https://github.com/denismakogon/gocv-alpine](https://github.com/denismakogon/gocv-alpine).
We have a special installation for the Raspberry Pi that includes some hardware optimizations. You use `make` to install OpenCV 4.9.0 with the handy `Makefile` included with this repo. If you already have installed OpenCV, you do not need to do so again. The installation performed by the `Makefile` is minimal, so it may remove OpenCV options such as Python or Java wrappers if you have already installed OpenCV some other way.
By default, pkg-config is used to determine the correct flags for compiling and linking OpenCV. This behavior can be disabled by supplying `-tags customenv` when building/running your application. When building with this tag you will need to supply the CGO environment variables yourself.
Please note that you will need to run these 3 lines of code one time in your current session in order to build or run the code, in order to setup the needed ENV variables. Once you have done so, you can execute code that uses GoCV with your custom environment like this:
Download and run the MinGW-W64 compiler installer from [https://sourceforge.net/projects/mingw-w64/files/Toolchains%20targetting%20Win32/Personal%20Builds/mingw-builds/8.1.0/](https://sourceforge.net/projects/mingw-w64/files/Toolchains%20targetting%20Win32/Personal%20Builds/mingw-builds/8.1.0/).
Choose the options for "posix" threads, and for "seh" exceptions handling, then install to the default location `c:\Program Files\mingw-w64\x86_64-8.1.0-posix-seh-rt_v6-rev0`.
Download and install CMake [https://cmake.org/download/](https://cmake.org/download/) to the default location. CMake installer will add CMake to your system path.
By default, OpenCV is expected to be in `C:\opencv\build\install\include`. This behavior can be disabled by supplying `-tags customenv` when building/running your application. When building with this tag you will need to supply the CGO environment variables yourself.
Due to the way OpenCV produces DLLs, including the version in the name, using this method is required if you're using a different version of OpenCV.
Please note that you will need to run these 3 lines of code one time in your current session in order to build or run the code, in order to setup the needed ENV variables. Once you have done so, you can execute code that uses GoCV with your custom environment like this:
Since memory allocations for images in GoCV are done through C based code, the go garbage collector will not clean all resources associated with a `Mat`. As a result, any `Mat` created _must_ be closed to avoid memory leaks.
To ease the detection and repair of the resource leaks, GoCV provides a `Mat` profiler that records when each `Mat` is created and closed. Each time a `Mat` is allocated, the stack trace is added to the profile. When it is closed, the stack trace is removed. See the [runtime/pprof documentation](https://golang.org/pkg/runtime/pprof/#Profile).
In order to include the MatProfile custom profiler, you MUST build or run your application or tests using the `-tags matprofile` build tag. For example:
We can see that this program would leak memory. As it exited, it had one `Mat` that was never closed. The stack trace points to exactly which line the allocation happened on (line 11, the `gocv.NewMat()`).
Furthermore, if the program is a long running process or if GoCV is being used on a web server, it may be helpful to install the HTTP interface )). For example:
This will leak a `Mat` once per second. You can see the current profile count and stack traces by going to the installed HTTP debug interface: [http://localhost:6060/debug/pprof/gocv.io/x/gocv.Mat](http://localhost:6060/debug/pprof/gocv.io/x/gocv.Mat?debug=1).
Please take a look at our [CONTRIBUTING.md](./CONTRIBUTING.md) document to understand our contribution guidelines.
Then check out our [ROADMAP.md](./ROADMAP.md) document to know what to work on next.
## Why this project exists
The [https://github.com/go-opencv/go-opencv](https://github.com/go-opencv/go-opencv) package for Go and OpenCV does not support any version above OpenCV 2.x, and work on adding support for OpenCV 3 had stalled for over a year, mostly due to the complexity of [SWIG](http://swig.org/). That is why we started this project.
The GoCV package uses a C-style wrapper around the OpenCV 4 C++ classes to avoid having to deal with applying SWIG to a huge existing codebase. The mappings are intended to match as closely as possible to the original OpenCV project structure, to make it easier to find things, and to be able to figure out where to add support to GoCV for additional OpenCV image filters, algorithms, and other features.
For example, the [OpenCV `videoio` module](https://github.com/opencv/opencv/tree/master/modules/videoio) wrappers can be found in the GoCV package in the `videoio.*` files.
This package was inspired by the original https://github.com/go-opencv/go-opencv project, the blog post https://medium.com/@peterleyssens/using-opencv-3-from-golang-5510c312a3c and the repo at https://github.com/sensorbee/opencv thank you all!
