Har*_*ris 5 java android opencv
将提供2张图像.我们需要找出它们之间的差异并突出它们.
到目前为止,我已经在JAVA中看到了这个解决方案,但由于android不支持BufferedImage,我无法继续进行.我已接近比较2位图的像素,但面临未来的问题.
我也尝试比较两个位图的像素,但它突出显示所有非白色
void findDifference(Bitmap firstImage, Bitmap secondImage)
{
if (firstImage.getHeight() != secondImage.getHeight() && firstImage.getWidth() != secondImage.getWidth())
Toast.makeText(this, "Images size are not same", Toast.LENGTH_LONG).show();
boolean isSame = true;
for (int i = 0; i < firstImage.getWidth(); i++)
{
for (int j = 0; j < firstImage.getHeight(); j++)
{
if (firstImage.getPixel(i,j) == secondImage.getPixel(i,j))
{
}
else
{
differentPixels.add(new Pixel(i,j));
secondImage.setPixel(i,j, Color.YELLOW); //for now just changing difference to yello color
isSame = false;
}
}
}
imgOutput.setImageBitmap(secondImage);
}
Run Code Online (Sandbox Code Playgroud)
提前致谢.
在数字色度计的帮助下,我发现在视觉上看起来相似的图像的颜色之间存在非常微小的差异,这与@taarraas建议的非常相似.所以我拿了一个阈值并像这样解决了.
private static final int threashold = 10;
void findDifference(Bitmap firstImage, Bitmap secondImage)
{
if (firstImage.getHeight() != secondImage.getHeight() || firstImage.getWidth() != secondImage.getWidth())
Toast.makeText(this, "Images size are not same", Toast.LENGTH_LONG).show();
boolean isSame = true;
for (int i = 0; i < firstImage.getWidth(); i++)
{
for (int j = 0; j < firstImage.getHeight(); j++)
{
int pixel = firstImage.getPixel(i,j);
int redValue = Color.red(pixel);
int blueValue = Color.blue(pixel);
int greenValue = Color.green(pixel);
int pixel2 = secondImage.getPixel(i,j);
int redValue2 = Color.red(pixel2);
int blueValue2 = Color.blue(pixel2);
int greenValue2 = Color.green(pixel2);
if (Math.abs(redValue2 - redValue) + Math.abs(blueValue2 - blueValue) + Math.abs(greenValue2 - greenValue) <= threashold)
// if (firstImage.getPixel(i,j) == secondImage.getPixel(i,j))
{
}
else
{
differentPixels.add(new Pixel(i,j));
secondImage.setPixel(i,j, Color.YELLOW); //for now just changing difference to yello color
isSame = false;
}
}
}
imgOutput.setImageBitmap(secondImage);
}
Run Code Online (Sandbox Code Playgroud)