How to do image processing in Go?

As an efficient programming language, Go also has good performance in the field of image processing. Although Go's own standard library does not provide specialized image processing-related APIs, there are some excellent third-party libraries for us to use, such as GoCV, ImageMagick, and GraphicsMagick. This article will focus on using GoCV for image processing.

GoCV is a Go language binding library that is highly dependent on OpenCV. Its API design is very similar to Python's opencv-python and C's OpenCV, so it is easy to learn and get started. Used for processing images, videos, cameras and other tasks. Below we will introduce the implementation of several commonly used image processing tasks.

1. Image loading and saving

Before image processing, you need to read the image in and save the processed image. GoCV provides many functions to help us achieve this process. The following is an example of loading and storing an image:

package main

import (
    "fmt"
    "gocv.io/x/gocv"
)

func main() {
    img := gocv.IMRead("test.jpg", gocv.IMReadColor)

    if img.Empty() {
        fmt.Println("读取图像失败")
        return
    }

    gocv.IMWrite("out.jpg", img)
}

In this example, the IMRead function is used to read an image in JPG format, and the second parameter specifies the read image The method of conversion is required, where gocv.IMReadColor indicates that the read image needs to be converted into a color image. Then we determine whether the reading is successful. If the read image is empty, then the reading fails. Finally, use the IMWrite function to save the image to the specified location. The image saved here is also in JPG format.

2. Image scaling

Image scaling is a very common task in image processing. Shrinking an image can be used to reduce image size and speed up calculations, while enlarging an image can be used to enhance image details. GoCV provides the Resize function to implement image scaling operations. The following is a simple example of scaling an image:

package main

import (
    "gocv.io/x/gocv"
)

func main() {
    img := gocv.IMRead("test.jpg", gocv.IMReadColor)
    dst := gocv.NewMat()

    gocv.Resize(img, &dst, image.Point{}, 0.5, 0.5, gocv.InterpolationDefault)

    gocv.IMWrite("out.jpg", dst)
}

In this example, we first use IMRead The function reads an image, and then uses the NewMat function to create a Mat object with the same size as the original image. The Resize function is used to reduce the original image to half, and finally use IMWrite to save the processed image to the specified location.

3. Image cropping

Image cropping can be used to perform local processing on images, and can play a very important role in extracting areas of interest, cropping useless information, and extracting target objects. important role. GoCV provides the ROI function to implement image cropping operations. The following is a simple image cropping example:

package main

import (
    "gocv.io/x/gocv"
)

func main() {
    img := gocv.IMRead("test.jpg", gocv.IMReadColor)
    dst := img.Region(gocv.NewRect(50, 50, 200, 200))

    gocv.IMWrite("out.jpg", dst)
}

In this example, we first use IMRead The function reads an image and then extracts a region of interest from it using the Region function. Here gocv.NewRect(50, 50, 200, 200) means that the cropped area of ​​interest is a rectangle with a length of 200 pixels, a width of 200 pixels, and the upper left corner coordinates are (50, 50) . Finally, use IMWrite to save the processed image to the specified location.

4. Image filtering

Image filtering can be used to remove image noise, smooth images and other operations. GoCV also provides many filter functions for us to use, including GaussianBlur, MedianBlur, BilateralFilter, etc. The following is an example of using Gaussian filtering:

package main

import (
    "gocv.io/x/gocv"
)

func main() {
    img := gocv.IMRead("test.jpg", gocv.IMReadGrayScale)
    dst := gocv.NewMat()

    gocv.GaussianBlur(img, &dst, image.Point{X: 5, Y: 5}, 0, 0, gocv.BorderDefault)

    gocv.IMWrite("out.jpg", dst)
}

In this example, we use the IMRead function to load a grayscale image, and then use the NewMat function to create A Mat object with the same dimensions as the original image. The Gaussian filter function GaussianBlur is used here, and the second parameter is the Mat object of the output result. The third parameter image.Point{X: 5, Y:5} represents the template size used when filtering, here is a rectangle with a length of 5 pixels and a width of 5 pixels. Finally, use IMWrite to save the processed image to the specified location.

5. Image segmentation

Image segmentation is an important image processing task. It can be used for tasks such as separating target objects and preprocessing data to generate specific features. GoCV provides the Canny function for edge detection, which can be used to implement simple image segmentation. The following is an example of using the Canny function:

package main

import (
    "gocv.io/x/gocv"
)

func main() {
    img := gocv.IMRead("test.jpg", gocv.IMReadGrayScale)
    dst := gocv.NewMat()

    gocv.Canny(img, &dst, 100, 200)

    gocv.IMWrite("out.jpg", dst)
}

In this example, we use the IMRead function to load a grayscale image, and then use the NewMat function to create A Mat object with the same dimensions as the original image. The Canny edge detection function Canny is used here, and the second parameter is the Mat object of the output result. The third and fourth parameters 100, 200 represent the minimum and maximum thresholds respectively, which can be adjusted according to actual problems. Finally, use IMWrite to save the processed image to the specified location.

The above are how some common image processing tasks are implemented in the Go language. GoCV provides many excellent image processing functions and is well unified with other libraries in the Python and C fields. The entry barrier is low, so it is very suitable for beginners to learn and use.

The above is the detailed content of How to do image processing in Go?. For more information, please follow other related articles on the PHP Chinese website!