Java 调整图像中心

Question

I was looking something else up the other day and feel like I remember running across a way to resize an image but keeping an area along the sides of the image unaltered so that the rounded edges of the images will always look the same no matter the length of the scaled image. After some more googling I cannot find what I remember seeing so I would like to know if this is possible and if anyone has and info.

Specifically what I have is a custom border with what is meant to look like a hanging tab with a label on it. The image is square on top and rounded on the bottom edges. I would like to keep the bottom rounded corners no matter the length of the labels on it but stretch the center ~90% to accommodate larger labels. I assume it would be in this code somewhere where i can enter some insets? in the resizing method.

titleLabel = new JLabel(title){
            public void paint(Graphics g){
                g.drawImage(tabImg, 0, 0, this.getSize().width, this.getSize().height, null);
                g.drawString(this.getText(), 1, this.getFont().getSize());
            }
        };

Answer 1

OK, given the lack of accepted method to do this I have come up with this solution and am posting the code for anyone else who would like to do something similar. This code will take an image and divide it into 9 sections. The 4 corners will be left alone. the middles of the 4 edges will be stretched or compressed along the edge. The center section will be stretched or compressed in both directions. Of course the point of this class for me was to compress a larger image with rounded corners but keep the rounded corners which nearly disappeared when the image was simply scaled down. Obviously this will do little good with an image like a picture but for components with custom painting and rounded edges this seems to be working good.

There is no constructor for this class you can just simply call for an altered image. the use would be

StretchedImage.stretch(image, new Insets(t,l,b,r), new Dimension(w,h), BufferedImage.TYPE_INT_ARGB); 

This will return an image stretched to the desired dimension with the corners remaining the same and the sides only being modified in 1 dimension using the insets parameter to determine the amount around the edge that is modified in a single dimension. There is probably a tricky way to iterate through the lists of images but this way I could better see what was going on.

public class StretchedImage{

public static Image stretch(Image image, Insets ins, Dimension dim, int hints){
    //debugcode
    //System.out.println(dim); 

    //load image into bufferedImage
    BufferedImage bi = toBufferedImage(image, hints); 

    //create 2d bufferedImage array to hold the 9 images
    Image[][] img = new Image[3][3]; 

    //split Image into 9 subsections
    img[0][0] = bi.getSubimage(0, 0, ins.left, ins.top);
    img[0][1] = bi.getSubimage(ins.left, 0, bi.getWidth() - ins.left - ins.right, ins.top);
    img[0][2] = bi.getSubimage(bi.getWidth() - ins.right, 0, ins.right, ins.top);
    img[1][0] = bi.getSubimage(0, ins.top, ins.left, bi.getHeight() - ins.top - ins.bottom);
    img[1][1] = bi.getSubimage(ins.left, ins.top, bi.getWidth() - ins.left - ins.right, bi.getHeight() - ins.top - ins.bottom);
    img[1][2] = bi.getSubimage(bi.getWidth() - ins.right, ins.top, ins.right, bi.getHeight() - ins.top - ins.bottom);
    img[2][0] = bi.getSubimage(0, bi.getHeight() - ins.bottom, ins.left, ins.bottom);
    img[2][1] = bi.getSubimage(ins.left, bi.getHeight() - ins.bottom, bi.getWidth() - ins.left - ins.right, ins.bottom);
    img[2][2] = bi.getSubimage(bi.getWidth() - ins.right, bi.getHeight() - ins.bottom, ins.right, ins.bottom);

    //determine the width and height of the sections that will be stretched
    //only the center section will have both dimensions changed
    int w = dim.width - ins.left - ins.right;
    int h = dim.height - ins.top - ins.bottom;

    //Stretch the proper sections 
    img[0][1] = img[0][1].getScaledInstance(w, img[0][1].getHeight(null), hints);
    img[1][0] = img[1][0].getScaledInstance(img[1][0].getWidth(null), h, hints);
    img[1][1] = img[1][1].getScaledInstance(w, h, hints);
    img[1][2] = img[1][2].getScaledInstance(img[1][2].getWidth(null), h, hints);
    img[2][1] = img[2][1].getScaledInstance(w, img[2][1].getHeight(null), hints);

    //for loop is debug code
    //for(int i = 0; i < 3; i++){ 
    //  for(int j = 0; j < 3; j++){
    //      System.out.println(i + " " + j + " " + img[i][j].getWidth() + "," + img[i][j].getHeight());
    //  }
    //}

    //create a new bufferedImage to hold the final image
    BufferedImage finalImage = new BufferedImage(dim.width, dim.height, hints);
    Graphics g = finalImage.getGraphics();
    //draw the peices to the final image
    g.drawImage(img[0][0], 0, 0, null);
    g.drawImage(img[0][1], ins.left, 0, null);
    g.drawImage(img[0][2], dim.width - ins.right, 0, null);
    g.drawImage(img[1][0], 0, ins.top, null);
    g.drawImage(img[1][1], ins.left, ins.top, null);
    g.drawImage(img[1][2], dim.width - ins.right, ins.top, null);
    g.drawImage(img[2][0], 0, dim.height - ins.bottom, null);
    g.drawImage(img[2][1], ins.left, dim.height - ins.bottom, null);
    g.drawImage(img[2][2], dim.width - ins.right, dim.height - ins.bottom, null);   

    return (Image)finalImage;
}

// This method returns a buffered image with the contents of an image
public static BufferedImage toBufferedImage(Image image, int hints) {

    BufferedImage bi = new BufferedImage(image.getWidth(null), image.getHeight(null), hints);
    bi.getGraphics().drawImage(image, 0, 0, null);

    return bi;
}

}