I am unable to find a method to detect the text area in the document and apply a filter to the rest of the image to clear it from any noise. Please refer to this image. If I do apply a filter to the image anyway, the text doesn't remain visible anymore.
Is there an algorithm in MATLAB that can help me find the textarea and treat it separately?
You can greatly reduce the amount of noise by applying erosion, followed by a gaussian blur (σ=2; the image will be converted to grayscale) followed by conversion back to binary.
Erosion alone will reduce the amount of noise significantly:
After application of gaussian blur and re-conversion into binary, the noise will be further reduced. Note that very small text (like the sub-headline) will be also degraded:
All three needed operations seem to be already implemented in MatLab: See Erosion; Gaussian Blur and Thresholding. However, note that for the example automatic (histogram based) thresholding was used to determine the optimal threshold.
Related
There are many ways to implement math morph on binary image like imerode and imdilate. Its also used to detect different object/shape using this simple operations on binary image but the problem that i am facing right now is to apply this simple operation i.e erode, dilate and many on grey scale image with out convert them into binary image.
Selement = strel('disk',5);//disk type element used in morphology
erodeimage = imerode(image,selement);//this is only implement on binary image
Above code is for binary math morph how do i implement same concept on grey scale image.
Note: If your have any resources regarding gray scale math morph kindly provide it or provide useful link
There should be a mathematical morphology (MM) library in MatLab. MM operations on binary images are shown as example/illustration but are performed most of the time as gray level.
I think that the fastest C++ library is SMIL, and you can call it from MatLab. An other fast one in C is that one (optimized opening/closing in a single pass).
But if you want to understand a dilation in gray level, here is how it works: for a given pixel p you analyze the value of all the pixels in its neighborhood (defined by the structuring element), and you affect to p the highest value in the neighborhood. You do it for each pixel in your image. See the formula.
That's in fact a rank filter like the median, but instead of taking the median value, you take the max (or min for the erosion). Obviously that the basic definition and it exits faster algorithms, like the one developed in the library I pointed.
I have been trying to restore a noisy image on MATLAB. I started with an original grayscale image of mine and then I applied Gaussian noise. I then took the noisy image and applied a Gaussian smoothing filter. After applying the smoothing filter, I applied a Laplacian filter over the Gaussian Blurred image and got a black image with some "edges" showing. What I am confused about is what to do next. I tried using the imadd function on MATLAB and adding the Gaussian blurred image with output of the Laplacian filter, but my results are not as good as I thought they would be. The "restored" image is nowhere near as good as I thought it would be!
Am I doing this correctly?
#eigenchris basically nailed it right on the head, but I would like to elaborate some more on why we believe this is a bad idea. Blurring the image removes high frequency content (i.e. edges). If you try to apply a high-pass filter like the Laplacian to the low-pass result, you will probably not get anything at all.
Specifically, the high frequency components were removed when you Gaussian blurred the image, and so if you apply a high-pass filter to an image with high frequency components already removed, you will probably get an almost zero output.
The moral of this story is that you can't sharpen an already blurred image because it relies on high frequency information to facilitate the sharpening. You are essentially amplifying the high frequency content so that the edges stand out more, and hence it is a sharpened result.
One thing I could suggest is to perhaps look into deconvolution techniques, like the Wiener filter. The Wiener filter essentially tries to undo the effects performed by a filter done on an image.
One great example can be found on this MathWorks link: http://www.mathworks.com/help/images/examples/deblurring-images-using-a-wiener-filter.html
As such, blur the image to eliminate any noise, then reverse the blur with Wiener filtering so you can get an OK version of the original, then sharpen that reconstructed image.
Good luck!
I am a new to Matlab and I have a project that involves image processing.
I have a number of RGB images and I need to find a way to separate the out of focus from the in focus images. I do not need to correct the focus of the out of focus ones, I just need to find which are out of focus and remove them. I have done FFT2 to the image and then used the radial average of the image of the power spectrum to see if there is a difference between the in focus or out of focus but I do not see a difference between the two.
I decided to use the gradient of the image
[gradx,grady]=gradient(image)
and then take the magnitude
new_image=sqrt((gradx.^2)+(grady.^2))
and try to do the FFT2 using the new_image now instead of the image. The power spectrum does not look like what I expect so I am not sure if I should do the FFT2 on the new_image of the gradx and grady separately. Has anyone have any thoughts about whether this is the right way to do this?
I was also thinking that instead of using the gradient to use a Sobel mask
mask=fspecial('sobel')
mask_x=imfilter(image,mask)
mask_y=imfilter(image,mask')
new_image=sqrt((mask_x.^2)+(mask_y.^2))
and then do FFT2 in the new_image but again the power spectrum is not right. I expect it to start from zero and instead it starts from the highest value and drops exponentially.
Has anyone tried to classify images using this method? Thank you for reading.
A DCT, instead of an FFT/DFT, will get rid of any high frequency discontinuities between the opposite edges of your images.
Hi I'm attempting to filter an image with 4 objects inside using MatLab. My first image had a black background with white objects so it was clear to me to filter each image out by finding these large white sections using BW Label and separating them from the image.
The next image has noise in it though. Now I have an image with white lines running through my objects and they are actually connected to each other now. How could I filter out these lines in MatLab? What about Salt and pepper noise? Are there MatLab functions that can do this?
Filtering noise can be done in several ways. A typical noise filtering procedure will be something like threshold>median filtering>blurring>threshold. However, information regarding the type of noise can be very important for proper noise filtration. For examples, since you have lines in your image you can try to use a Hough transform to detect them and take them out of the play (or houghlines). Another approach can be to implement RANSAC. For salt & pepper type of noise, one should use medfilt2 with a proper window size that captures the noise characteristics (for example 3x3 window will deal well with noise fluctuations that are 1 pixel big...).
If you can live with distorting the objects a bit, you can use a closing (morphological) filter with a bit of contrast stretching. You'll need the image processing toolbox, but here's the general idea.
Blur to kill the lines otherwise the closing filter will erase your objects. You can use fspecial to create a Gaussian filter and imfilter to apply it
Apply the closing filter to the image using imclose with a mask that's bigger then your noise, but smaller then the object pieces (I used a 3x3 diamond in my example).
Threshold your image using im2bw so that every pixel gets turned to pure black or pure white based
I've attached an example I had to do for a school project. In my case, the background was white and objects black and I stretched between the erosion and dilation. You can't really see the gray after the erosion, but it was there (hence the necessity for thresholding).
You can of course directly do the closing (erosion followed by dilation) and then threshold. Notice how this filtering distorts the objects.
FYI usually salt-and-pepper noise is cleaned up with a moving average filter, but that will leave the image grayscale. For my project, I needed a pure black and white (for BW Label) and the morphological filters worked great to completely obliterate the noise.
I've scanned an old photo with paper texture pattern and I would like to remove the texture as much as possible without lowering the image quality. Is there a way, probably using Image Processing toolbox in MATLAB?
I've tried to apply FFT transformation (using Photoshop plugin), but I couldn't find any clear white spots to be paint over. Probably the pattern is not so regular for this method?
You can see the sample below. If you need the full image I can upload it somewhere.
Unfortunately, you're pretty much stuck in the spatial domain, as the pattern isn't really repetitive enough for Fourier analysis to be of use.
As #Jonas and #michid have pointed out, filtering will help you with a problem like this. With filtering, you face a trade-off between the amount of detail you want to keep and the amount of noise (or unwanted image components) you want to remove. For example, the median filter used by #Jonas removes the paper texture completely (even the round scratch near the bottom edge of the image) but it also removes all texture within the eyes, hair, face and background (although we don't really care about the background so much, it's the foreground that matters). You'll also see a slight decrease in image contrast, which is usually undesirable. This gives the image an artificial look.
Here's how I would handle this problem:
Detect the paper texture pattern:
Apply Gaussian blur to the image (use a large kernel to make sure that all the paper texture information is destroyed
Calculate the image difference between the blurred and original images
EDIT 2 Apply Gaussian blur to the difference image (use a small 3x3 kernel)
Threshold the above pattern using an empirically-determined threshold. This yields a binary image that can be used as a mask.
Use median filtering (as mentioned by #Jonas) to replace only the parts of the image that correspond to the paper pattern.
Paper texture pattern (before thresholding):
You want as little actual image information to be present in the above image. You'll see that you can very faintly make out the edge of the face (this isn't good, but it's the best I have time for). You also want this paper texture image to be as even as possible (so that thresholding gives equal results across the image). Again, the right hand side of the image above is slightly darker, meaning that thresholding it well will be difficult.
Final image:
The result isn't perfect, but it has completely removed the highly-visible paper texture pattern while preserving more high-frequency content than the simpler filtering approaches.
EDIT
The filled-in areas are typically plain-colored and thus stand out a bit if you look at the image very closely. You could also try adding some low-strength zero-mean Gaussian noise to the filled-in areas to make them look more realistic. You'd have to pick the noise variance to match the background. Determining it empirically may be good enough.
Here's the processed image with the noise added:
Note that the parts where the paper pattern was removed are more difficult to see because the added Gaussian noise is masking them. I used the same Gaussian distribution for the entire image but if you want to be more sophisticated you can use different distributions for the face, background, etc.
A median filter can help you a bit:
img = imread('http://i.stack.imgur.com/JzJMS.jpg');
%# convert rgb to grayscale
img = rgb2gray(img);
%# apply median filter
fimg = medfilt2(img,[15 15]);
%# show
imshow(fimg,[])
Note that you may want to pad the image first to avoid edge effects.
EDIT: A smaller filter kernel than [15 15] will preserve image texture better, but will leave more visible traces of the filtering.
Well i have tried out a different approach using Anisotropc diffusion using the 2nd coefficient that operates on wider areas
Here is the output i got:
From what i can See from the Picture, the Noise has a relatively high Frequency Compared to the image itself. So applying a low Pass filter should work. Have a look at the Power spectrum abs(fft(...)) to determine the cutoff Frequency.