Cell segmentation - python

I'm newbie in computer vision. My goal is to distinguish individual cells on a set of pictures like this: Example
Basically, I blur whole image, find region maximum on it and use it like seed in watershed algorithm on distance tranfsform of threesholded blurred image. In fact I'm following tutorial which you can find here:
github/luispedro/python-image-tutorial
(sorry, can't post more than 2 links).
My problem is that some cells in my set have very distinguishable dark nucleus (which you can see on the example) and my algorithm produce results like this which are cleary wrong.
Of course it's possible to fix it by increasing strength of gaussian blur but it will merge some other cells toghether which is even worse.
What can be done to solve this problem? What are other possibilites if watershed just isn't situable for this case (keeping in mind that my set is pretty small and learning seems impossible)?

The watershed tends to over-segment if you don't use a watershed with markers.
Usually, we start with DNA/DAPI segmentation that is easy, and it provides the number of cells and the inner markers for the watershed.
If you blur the images, you smooth all the patterns. You should use an alternate sequential filter (opening / closing) in order to simplify each zone, and then try an ultimate eroded in order to find the number of inner seed for your watershed.

Related

Finding contour around a cluster of pixels

I have a set of images that look like this:
Using python need a way to find a contour around the yellow shape that ignores the isolated points and is not too complex. Something looking a bit like this :
I tried some methods such as the find_contours function from skimage,which gives this after keeping only the biggest contour:
which is not what I am looking for. A also tried active contour (snake) which had the problem of paying too much attention to isolated pixels. Is there a particular method that would help me in this situation ?
Thank you
Assuming the yellow blob is slightly different across your images, I recommend you look into either using Morphological Operations, or using Contour Approximation.
I've never used scikit-image, but it appears to have Morphological functionalities included.
You can take a look at this OpenCV tutorial for a quick guideline of the different operations.
But I think all you need is to use the "Opening" operation to preprocess your yellow shape; making it smoother and removing the random speckles.
Another approach is by approximating that contour you've extracted to make it smoother. For scikit-image, that is the measure.approximate_polygon function. Also another OpenCV tutorial for reference on how Contour Approximation works (the same algorithm as with scikit-image).

Smoothen edges of a non-binary image

I'm making some photo-editing tools in python using PIL (Python Imaging Library), and I was trying to make a program which converts a photo to its 'painted' version.
I've managed to make a program which converts a photo into its distinct colours, but the problem is that the algorithm I'm using is operating on every pixel, meaning that the resulting image has very jagged differences between colours.
Ideally, I'd like to smoothen out these edges, but I don't know how!
I've checked out this site for some help, but the method there produces quite different results to what I need.
My Starting Image:
My Image with Distinct Colours:
I would like to smoothen the edges in the image above.
Results of using the method which doesn't quite work:
As you can see, using the technique doesn't smoothen the edges into natural-looking curves; instead it creates jagged edges.
I know I should provide sample output, but suprisingly, I haven't actually got it, so I'll describe it as best as I can. Simply put, I want to smoothen the edges between the different colours.
I've seen something called a Gaussian blur, but I'm not quite sure as to how to apply it here as the answers I've seen always mention some sort of threshold, and are usually to do with binary images, so I don't think it can apply here.
Edge enhancement does the opposite of edge smoothing, so this is certainly not the tool you should use.
Unfortunately, there is little that you can do because edge smoothing will indeed smoothen the jaggies, but it will also destroy the true edges, resulting in a blurred image. Edge-preserving smoothing is also a dead-end.
You should have a look at the methods to extract the "cartoon part" of an image. There is a lot of literature on this topic, though often pretty sophisticated.
You can enhance the quality of your "Image with Distinct Colours" by applying a median filter with a radius of 2:
If you want to get "comic-like" dark edges, you can calculate the edges of the original image using a sobel filter, convert the edge map to grayscale, then multiply the resulting edge map with 2, inverse the map and add each non-white pixel of the edge map to the original image. This will result in:
This is of course only a starting point as the result leaves much to be desired, but it should give you a good idea about the basic concept.

Removing text while processing the image

I am working on an application where I need feature like Cam Scanner where document is to be detected in an image. For that I am using Canny Edge detection followed by Hough Transform.
The results look promising but the text in the document is creating issues as explained via images below:
Original Image
After canny edge detection
After hough transform
My issue lies in the third image, the text in original mage near the bottom has forced hough transform to detect the horizontal line(2nd cluster from bottom).
I know I can take the largest quadrilateral and that would work fine in most cases, but still I want to know any other ways where in this processing I can ignore the effect of text on the edges.
Any help would be appreciated.
I solved the issue of text with the help of median filter of size 15(square) in an image of 500x700.
Median filter doesn't affect the boundaries of the paper, but can help eliminate the text completely.
Using that I was able to get much more effective boundaries.
Another approach you could try is to use thresholding to find the paper boundaries. This would create a binary image. You can then examine the blobs of white pixels and see if any are large enough to be the paper and have the right dimensions. If it fits the criteria, you can find the min/max points of this blob to represent the paper.
There are several ways to do the thresholding, including iterative, otsu, and adaptive.
Also, for best results you may have to dilate the binary image to close the black lines in the table as shown in your example.

Smooth Boundary Extraction from a Binary Image

I'm trying to extract the boundary between two regions in an image programmatically. I've got the hard bits figured out, so that I have a binary image that contains the boundary and plenty of noise.
. Cropping the areas outside isn't an issue.
The boundary in the image is afflicted both by noise (bottom-left for example) and some areas of discontinuity. That means I can't simply select the shape based on one known pixel.
The problem left to me is pretty simple - I only really need to fill the gaps in the boundary and smooth it out, so that I am left with something smooth and continuous that I can extract afterwards. That doesn't sound like a particularly hard problem for images like this, but I'm completely lost. What algorithms or strategies could I possible use in order to turn this image into something useful?
The output I'm looking for is something that can be cropped to give .
An common practice is to use a gaussian blur. This will filter out the noise in the image depending on the intensity of the blur. At the bottom of the article there is a gif with a cat image showing what you want.
After that there are contour finding algos which could help you extract the boundaries as pixel chains

Robust detection of grid pattern in an image

I have written a program in Python which automatically reads score sheets like this one
At the moment I am using the following basic strategy:
Deskew the image using ImageMagick
Read into Python using PIL, converting the image to B&W
Calculate calculate the sums of pixels in the rows and the columns
Find peaks in these sums
Check the intersections implied by these peaks for fill.
The result of running the program is shown in this image:
You can see the peak plots below and to the right of the image shown in the top left. The lines in the top left image are the positions of the columns and the red dots show the identified scores. The histogram bottom right shows the fill levels of each circle, and the classification line.
The problem with this method is that it requires careful tuning, and is sensitive to differences in scanning settings. Is there a more robust way of recognising the grid, which will require less a-priori information (at the moment I am using knowledge about how many dots there are) and is more robust to people drawing other shapes on the sheets? I believe it may be possible using a 2D Fourier Transform, but I'm not sure how.
I am using the EPD, so I have quite a few libraries at my disposal.
First of all, I find your initial method quite sound and I would have probably tried the same way (I especially appreciate the row/column projection followed by histogramming, which is an underrated method that is usually quite efficient in real applications).
However, since you want to go for a more robust processing pipeline, here is a proposal that can probably be fully automated (also removing at the same time the deskewing via ImageMagick):
Feature extraction: extract the circles via a generalized Hough transform. As suggested in other answers, you can use OpenCV's Python wrapper for that. The detector may miss some circles but this is not important.
Apply a robust alignment detector using the circle centers.You can use Desloneux parameter-less detector described here. Don't be afraid by the math, the procedure is quite simple to implement (and you can find example implementations online).
Get rid of diagonal lines by a selection on the orientation.
Find the intersections of the lines to get the dots. You can use these coordinates for deskewing by assuming ideal fixed positions for these intersections.
This pipeline may be a bit CPU-intensive (especially step 2 that will proceed to some kind of greedy search), but it should be quite robust and automatic.
The correct way to do this is to use Connected Component analysis on the image, to segment it into "objects". Then you can use higher level algorithms (e.g. hough transform on the components centroids) to detect the grid and also determine for each cell whether it's on/off, by looking at the number of active pixels it contains.

Categories

Resources