I wrote an application where some dots are floating around and if i assign a dot a Point, it will move on this position. Now i want to load an image, convert it to a monochrome image (Only pure black or white pixels - no shades of gray) and make each dot floating to a position where its representing a black pixel.
I've already done loading and converting a image that way and extraceted the pixels as a 1 dimensional byte[]. I've managed to iterate though this array with the following code:
int stride = width * 4;
for (int y = 0; y < height; y++)
for (int x = 0; x < width; x++)
{
int index = y * stride + 4 * x;
// array[index] <- Red
// array[index + 1] <- Green
// array[index + 2] <- Blue
// array[index + 3] <- Alpha
}
The byte array holds every pixel with 4 bytes (RGBA). So the array length is ImageHeight*ImageWidth*4 bytes. Either a pixel is Black (0, 0, 0, 255) or White (255, 255, 255, 255).
My problem now is that i'm not able to correctly approximate the black areas of the image with just n dots. In most cases I will have much less floating dots than there are black pixels in the array. So what i need is a method that gives me a Point[] that contains n Points that will represent only the black areas of the image as good as possible. Can someone help me out?
Loop though the array and find the points that their red, green and blue are 0
to get the Black dots:
List<Point> blackPoints = new List<Points>()
for(int i=0; i<array.Length; i+=4)
if(array[i] == 0 && array[i+1] == 0 && array[i+2] ==0) //alpha is not important
{
int tmp = i / 4;
blackPoints.Add(new Point(tmp%width, tmp/width));
}
Create Methods, to get the weight of a pixel based on its own and neighbors colors, and also a Method to find the weight of a block:
public static class Exts
{
public static int Weight(this points ps, int x, int y, int width, int height)
{
int weight = 0;
for(int i=Math.Max(x - 1, 0); i<Math.Min(width, x+1))
for(int j= Math.Max(y-1, 0), j<Math.Min(height, y+1))
if(ps.Any(a => a.X == i && a.Y == j)) weight++;
return weight;
}
public static int BlockWeight(this Point[] ps, int x, int y)
{
return ps.Count(a => a.X <= x+2 && a.Y<= y+2);
}
}
Now loop through the bitmap, with blocks of nine pixels (3x3) and if a blocks wieght is more than half (in this case more than or equal to 5), select a the point in this block that has heighest weight, to represent the black point:
List<Point> result = new List<Point>();
for(int i=0; i<width; i+=3)
for(int j=0; j< height; j+=3)
if(points.BlockWeight(i,j) >= 5)
result.Add(ps.Where(a => a.X <= x+2 && a.Y<= y+2).OrderByDescending(a => a.Weight(i, j, width, height)).First())
Related
I did try to use this code in my application from this problem:
How to determine the background color of document when there are 3 options, using c# or imagemagick
private System.Drawing.Color CalculateAverageColor(Bitmap bm)
{
int width = bm.Width;
int height = bm.Height;
int red = 0;
int green = 0;
int blue = 0;
int minDiversion = 15; // drop pixels that do not differ by at least minDiversion between color values (white, gray or black)
int dropped = 0; // keep track of dropped pixels
long[] totals = new long[] { 0, 0, 0 };
int bppModifier = bm.PixelFormat == System.Drawing.Imaging.PixelFormat.Format24bppRgb ? 3 : 4; // cutting corners, will fail on anything else but 32 and 24 bit images
BitmapData srcData = bm.LockBits(new System.Drawing.Rectangle(0, 0, bm.Width, bm.Height), ImageLockMode.ReadOnly, bm.PixelFormat);
int stride = srcData.Stride;
IntPtr Scan0 = srcData.Scan0;
unsafe
{
byte* p = (byte*)(void*)Scan0;
for (int y = 0; y < height; y++)
{
for (int x = 0; x < width; x++)
{
int idx = (y * stride) + x * bppModifier;
red = p[idx + 2];
green = p[idx + 1];
blue = p[idx];
if (Math.Abs(red - green) > minDiversion || Math.Abs(red - blue) > minDiversion || Math.Abs(green - blue) > minDiversion)
{
totals[2] += red;
totals[1] += green;
totals[0] += blue;
}
else
{
dropped++;
}
}
}
}
int count = width * height - dropped;
int avgR = (int)(totals[2] / count);
int avgG = (int)(totals[1] / count);
int avgB = (int)(totals[0] / count);
return System.Drawing.Color.FromArgb(avgR, avgG, avgB);
}
The problem is when I add a bitmap, this happens:
Can someone help me figure out what exactly is happening?
While there is an issue, the bigger question is how to determine the issue?
You're pointing out where it crashes, but why would it crash there? It would only produce that error if count is zero.
int count = width * height - dropped;
Assuming the width and height are correct (and don't assume, put a breakpoint on this line and verify it!), this means dropped must be equal to width * height.
if (Math.Abs(red - green) > minDiversion || Math.Abs(red - blue) > minDiversion || Math.Abs(green - blue) > minDiversion)
...
else
{
dropped++;
}
This code means that if the pixel is a grayscale color (or close to it), then it will increment dropped. The idea is that it's only going to give you the average of the pixels with color, and avoid having grayscale pixels skew the results.
So, what's the solution? There are two possible answers, only you can pick the right one for your problem set:
if (Math.Abs(red - green) > minDiversion || Math.Abs(red - blue) > minDiversion || Math.Abs(green - blue) > minDiversion)
You could just get rid of this check and always add something to the totals values. This would give you a color in all cases, but how right that is depend on your needs.
int count = width * height - dropped;
if (count == 0)
{
return Color.Black;
}
Or, you could check for the case where you can't find a color and just return a default response (or, maybe throw an Exception).
Depends on your needs.
I should let everyone see this to understand the problem better
This is what i did in form1 constructor:
Bitmap bmp2 = new Bitmap(#"e:\result1001.jpg");
CropImageWhiteAreas.ImageTrim(bmp2);
bmp2.Save(#"e:\result1002.jpg");
bmp2.Dispose();
The class CropImageWhiteAreas:
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Drawing;
using System.Drawing.Imaging;
using System.Runtime.InteropServices;
namespace Test
{
class CropImageWhiteAreas
{
public static Bitmap ImageTrim(Bitmap img)
{
//get image data
BitmapData bd = img.LockBits(new Rectangle(Point.Empty, img.Size),
ImageLockMode.ReadOnly, PixelFormat.Format32bppArgb);
int[] rgbValues = new int[img.Height * img.Width];
Marshal.Copy(bd.Scan0, rgbValues, 0, rgbValues.Length);
img.UnlockBits(bd);
#region determine bounds
int left = bd.Width;
int top = bd.Height;
int right = 0;
int bottom = 0;
//determine top
for (int i = 0; i < rgbValues.Length; i++)
{
int color = rgbValues[i] & 0xffffff;
if (color != 0xffffff)
{
int r = i / bd.Width;
int c = i % bd.Width;
if (left > c)
{
left = c;
}
if (right < c)
{
right = c;
}
bottom = r;
top = r;
break;
}
}
//determine bottom
for (int i = rgbValues.Length - 1; i >= 0; i--)
{
int color = rgbValues[i] & 0xffffff;
if (color != 0xffffff)
{
int r = i / bd.Width;
int c = i % bd.Width;
if (left > c)
{
left = c;
}
if (right < c)
{
right = c;
}
bottom = r;
break;
}
}
if (bottom > top)
{
for (int r = top + 1; r < bottom; r++)
{
//determine left
for (int c = 0; c < left; c++)
{
int color = rgbValues[r * bd.Width + c] & 0xffffff;
if (color != 0xffffff)
{
if (left > c)
{
left = c;
break;
}
}
}
//determine right
for (int c = bd.Width - 1; c > right; c--)
{
int color = rgbValues[r * bd.Width + c] & 0xffffff;
if (color != 0xffffff)
{
if (right < c)
{
right = c;
break;
}
}
}
}
}
int width = right - left + 1;
int height = bottom - top + 1;
#endregion
//copy image data
int[] imgData = new int[width * height];
for (int r = top; r <= bottom; r++)
{
Array.Copy(rgbValues, r * bd.Width + left, imgData, (r - top) * width, width);
}
//create new image
Bitmap newImage = new Bitmap(width, height, PixelFormat.Format32bppArgb);
BitmapData nbd
= newImage.LockBits(new Rectangle(0, 0, width, height),
ImageLockMode.WriteOnly, PixelFormat.Format32bppArgb);
Marshal.Copy(imgData, 0, nbd.Scan0, imgData.Length);
newImage.UnlockBits(nbd);
return newImage;
}
}
}
I also tried before it Peter solution.
In both the result is(This is a screenshot of my facebook after uploaded the image) still the white areas around:
You can the rectangle around the image i just uploaded and see what i mean by white area around.
If I understand correctly, you have found a sample code snippet that uses LockBits(), but you are not sure how it works or how to modify it to suit your specific need. So I will try to answer from that perspective.
First, a wild guess (since you didn't include the implementation of the LockBitmap class you're using in the first example): the LockBitmap class is some kind of helper class that is supposed to encapsulate the work of calling LockBits() and using the result, including providing versions of GetPixel() and SetPixel() which are presumably much faster than calling those methods on a Bitmap object directly (i.e. access the buffer obtained by calling LockBits()).
If that's the case, then modifying the first example to suit your need is probably best:
public void Change(Bitmap bmp)
{
Bitmap newBitmap = new Bitmap(bmp.Width, bmp.Height, bmp.PixelFormat);
LockBitmap source = new LockBitmap(bmp),
target = new LockBitmap(newBitmap);
source.LockBits();
target.LockBits();
Color white = Color.FromArgb(255, 255, 255, 255);
for (int y = 0; y < source.Height; y++)
{
for (int x = 0; x < source.Width; x++)
{
Color old = source.GetPixel(x, y);
if (old != white)
{
target.SetPixel(x, y, old);
}
}
}
source.UnlockBits();
target.UnlockBits();
newBitmap.Save("d:\\result.png");
}
In short: copy the current pixel value to a local variable, compare that value to the white color value, and if it is not the same, go ahead and copy the pixel value to the new bitmap.
Some variation on the second code example should work as well. The second code example does explicitly what is (I've assumed) encapsulated inside the LockBitmap class that the first code example uses. If for some reason, the first approach isn't suitable for your needs, you can follow the second example.
In that code example you provide, most of the method there is just handling the "grunt work" of locking the bitmap so that the raw data can be accessed, and then iterating through that raw data.
It computes the oRow and nRow array offsets (named for "old row" and "new row", I presume) based on the outer y loop, and then accesses individual pixel data by computing the offset within a given row based on the inner x loop.
Since you want to do essentially the same thing, but instead of converting the image to grayscale, you just want to selectively copy all non-white pixels to the new bitmap, you can (should be able to) simply modify the body of the inner x loop. For example:
byte red = oRow[x * pixelSize + 2],
green = oRow[x * pixelSize + 1],
blue = oRow[x * pixelSize];
if (red != 255 || green != 255 || blue != 255)
{
nRow[x * pixelSize + 2] = red;
nRow[x * pixelSize + 1] = green;
nRow[x * pixelSize] = blue;
}
The above would entirely replace the body of the inner x loop.
One caveat: do note that when using the LockBits() approach, knowing the pixel format of the bitmap is crucial. The example you've shown assumes the bitmaps are in 24 bpp format. If your own bitmaps are in this format, then you don't need to change anything. But if they are in a different format, you'll need to adjust the code to suit that. For example, if your bitmap is in 32 bpp format, you need to pass the correct PixelFormat value to the LockBits() method calls, and then set pixelSize to 4 instead of 3 as the code does now.
Edit:
You've indicated that you would like to crop the new image so that it is the minimize size required to contain all of the non-white pixels. Here is a version of the first example above that should accomplish that:
public void Change(Bitmap bmp)
{
LockBitmap source = new LockBitmap(bmp);
source.LockBits();
Color white = Color.FromArgb(255, 255, 255, 255);
int minX = int.MaxValue, maxX = int.MinValue,
minY = int.MaxValue, maxY = int.MinValue;
// Brute-force scan of the bitmap to find image boundary
for (int y = 0; y < source.Height; y++)
{
for (int x = 0; x < source.Width; x++)
{
if (source.GetPixel(x, y) != white)
{
if (x < minX) minX = x;
if (x > maxX) maxX = x;
if (y < minY) minY = y;
if (y > maxY) maxY = y;
}
}
}
Bitmap newBitmap = new Bitmap(maxX - minx + 1, maxY - minY + 1, bmp.PixelFormat);
LockBitmap target = new LockBitmap(newBitmap);
target.LockBits();
for (int y = 0; y < target.Height; y++)
{
for (int x = 0; x < target.Width; x++)
{
target.SetPixel(x, y, source.GetPixel(x + minX, y + minY));
}
}
source.UnlockBits();
target.UnlockBits();
newBitmap.Save("d:\\result.png");
}
This example includes an initial scan of the original bitmap, after locking it, to find the minimum and maximum coordinate values for any non-white pixel. Having done that, it uses the results of that scan to determine the dimensions of the new bitmap. When copying the pixels, it restricts the x and y loops to the dimensions of the new bitmap, adjusting the x and y values to map from the location in the new bitmap to the given pixel's original location in the old one.
Note that since the initial scan determines where the non-white pixels are, there's no need to check again when actually copying the pixels.
There are more efficient ways to scan the bitmap than the above. This version simply looks at every single pixel in the original bitmap, keeping track of the min and max values for each coordinate. I'm guessing this will be fast enough for your purposes, but if you want something faster, you can change the scan so that it scans for each min and max in sequence:
Scan each row from y of 0 to determine the first row with a non-white pixel. This is the min y value.
Scan each row from y of source.Height - 1 backwards, to find the max y value.
Having found the min and max y values, now scan the columns from x of 0 to find the min x and from source.Width - 1 backwards to find the max x.
Doing it that way involves a lot more code and is probably harder to read and understand, but would involve inspecting many fewer pixels in most cases.
Edit #2:
Here is a sample of the output of the second code example:
Note that all of the white border of the original bitmap (shown on the left side) has been cropped out, leaving only the smallest subset of the original bitmap that can contain all of the non-white pixels (shown on the right side).
Given 2 consecutive frames, how can I search for pixels that changed?
I tried the following:
if (old != null)
{
for (int i = 0; i < b.Width; i++)
{
for (int j = 0; j < b.Height; j++)
{
if (!b.GetPixel(i, j).Equals(old.GetPixel(i, j)))
s.SetPixel(i, j, b.GetPixel(i, j));
else
s.SetPixel(i, j, Color.White);
}
}
}
Where "old" is the previous frame and "s" is the new frame. The code basically paints the pixels that didn't change in white.
But since the webcam produces a very low quality frame almost all of the pixels change.
How can I eliminate the pixels that didn't change "greatly"?
A very basic approach is to convert your Color pixel to an 0 - 255 based grey value.
So you can compare your pixels as an integer and make some delta error difference.
Consider this method which convert a color to a integer grayscale value
private static int GreyScaleRange(Color originalColor)
{
return (int)((originalColor.R * .3) + (originalColor.G * .59)
+ (originalColor.B * .11));
}
So instead of doing equal function, you should do
int deltadifference = 5 ;
if (Math.abs((GreyScaleRange(b.GetPixel(i, j)) - (GreyScaleRange(old.GetPixel(i, j)) > deltadifference)
s.SetPixel(i, j, b.GetPixel(i, j));
else
s.SetPixel(i, j, Color.White);
lets say that i have this image:
http://srv2.jpg.co.il/9/51c614f7c280e.png
i want to get the white ball position(x,y),
this is a verey big image, then i cut the image by rectangle.(because when the image is smaller everything is faster),
the results:
http://srv2.jpg.co.il/1/51c616787a3fa.png
now i want to track the white ball position by his color(white=rbg(255,255,255)),
my code:
Public Function GetBallPosition(ByRef HaxScreenOnly As Bitmap) As Point
For y = 0 To HaxScreenOnly.Height - 1
For x = 0 To HaxScreenOnly.Width - 1
If HaxScreenOnly.GetPixel(x, y) = Color.FromArgb(0, 0, 0) Then
If HaxScreenOnly.GetPixel(x + 8, y) = Color.FromArgb(0, 0, 0) And HaxScreenOnly.GetPixel(x + 8, y + 3) = Color.FromArgb(255, 255, 255) Then
Return New Point(x, y)
End If
End If
Next
Next
Return New Point(0, 0)
End Function
If the color of the current pixel is black and the color of the current pixel(x+8,y+3) is white then this is the ball
it's working...but its verey slow, something like 200 miliseconds to track the ball position.
this is not fast enough.
there is faster way to track the white ball(C# or VB.net)?
Finally, I have you a solution for you. Calling GetPixel is a costly process, but you use Bitmap.LockBits and manipulate / access the image data from a pointer. I took the LockBitmap class from this article.
I checked on the performance from what I was getting previously, which was exactly like you mentioned, around 200ms~.
Here is a picture of the result using LockBitmap, rescanning the image continuously with the optimized code!
static void Main(string[] args)
{
byte[] data = new WebClient().DownloadData("http://srv2.jpg.co.il/1/51c616787a3fa.png");
Image image = Image.FromStream(new MemoryStream(data));
LockBitmap bitmap = new LockBitmap(new Bitmap(image));
// this essentially copies the data into memory and copies from a pointer to an array
bitmap.LockBits();
Color black = Color.FromArgb(0, 0, 0);
Color white = Color.FromArgb(255, 255, 255);
Stopwatch stopwatch = Stopwatch.StartNew();
for (int y = 0; y < bitmap.Height; y++)
{
for (int x = 0; x < bitmap.Width; x++)
{
// GetPixel is a nice abstraction the author in the Article created so we don't have to do any of the gritty stuff.
if (bitmap.GetPixel(x, y) == black)
{
if (bitmap.GetPixel(x + 8, y) == black && bitmap.GetPixel(x + 8, y + 3) == white)
{
Console.WriteLine("White Ball Found in {0}", stopwatch.Elapsed.ToString());
break;
}
}
}
}
bitmap.UnlockBits(); // copies the data from the array back to the original pointer
Console.Read();
}
Hope this helps, it was certainly an interesting read for me.
Update:
As mentioned by King, I was able to further reduce the timing for you based on the algorithm improvement. So we've gone from O(n) to O(n log) time complexity (I think).
for (int y = 0; y < bitmap.Height; y += 3) // As we know the radius of the ball
{
for (int x = 0; x < bitmap.Width; x += 3) // We can increase this
{
if (bitmap.GetPixel(x, y) == black && bitmap.GetPixel(x, y + 3) == white)
{
Console.WriteLine("White Ball Found ({0},{1}) in {2}", x, y, stopwatch.Elapsed.ToString());
break;
}
}
}
i make region growing algorithm for my project
this is my algorithm
(my picture have been greyscale before it)
1. get value pixel (0,0) for seed pixel
2. compare value seed pixel with one neighbor pixel
3. if value of no.3 less than treshold (T), go to next pixel and go to no.2
4. if value of no.3 more than treshold (T), change pixel to white(also for next 10 pixel), and get new seed value pixel.
my goal is my picture segmented with white line
this is my code
private void button4_Click(object sender, EventArgs e)
{
// GDI+ still lies to us - the return format is BGR, NOT RGB.
BitmapData bmData = RImage.LockBits(new Rectangle(0, 0, RImage.Width, RImage.Height), ImageLockMode.ReadWrite, PixelFormat.Format24bppRgb);
int stride = bmData.Stride;
System.IntPtr Scan0 = bmData.Scan0;
unsafe
{
byte* p = (byte*)(void*)Scan0;
int nOffset = stride - RImage.Width * 3;
for (int y = 0; y < RImage.Height; ++y)
{
for (int x = 0; x < RImage.Width; ++x)
{
//every new line of x must new seed
if (x == 0)
{
//getting new value seed pixel
seedR = p[x];
seedG = p[x+1];
seedB = p[x+2];
}
//compare value of seed pixel and pixel scan
if ((seedR - p[x] >= tred) || (p[x] - seedR >= tred))
{
//make white line with change value of pixel
for (int i=1; i <= 5; ++i)
{
p[x] = p[x + 1] = p[x + 2] = 0;
x++;
}
//getting new value of seed pixel
seedR = p[x];
seedG = p[x + 1];
seedB = p[x + 2];
}
p += 3;
}
p += nOffset;
}
}
RImage.UnlockBits(bmData);
}
my problem is my image become white in 1/3 of image
what must i doing for "region growing" ??
thx
I've left some questions about your algorithm in the comments, but as I was writing them I realized that what you're trying to do may not be image segmentation at all.
my goal is my picture segmented with white line
Do you mean you want something like this:
If yes, then what you're interested in isn't image segmentation, it's edge detection. If you want to implement something like that, then have a read about convolution as well.