For my bachelor thesis I need to analyse images taken in the ocean to count and measure the size of water particles.
my problem:
besides the wanted water particles, the images show hexagonal patches all over the image in:
- different sizes
- not regular shape
- different greyscale values
(Example image below!)
It is clear that these patches will falsify my image analysis concerning the size and number of particles.
For this reason this patches need to be detected and deleted somehow.
Since it will be just a little part of the work in my thesis, I don't want to spend much time in it and already tried classic ways like: (imageJ)
playing with the threshold (resulting in also deleting wanted water particles)
analyse image including the hexagonal patches and later sort out the biggest areas (the hexagonal patches have quite the biggest areas, but you will still have a lot of haxagons)
playing with filters: using gaussian filter on a duplicated image and subtract the copy from the original deletes many patches (in reducing the greyscale value) but also deletes little wanted water particles and so again falsifies the result
a more complicated and time consuming solution would be to use a implemented library in for example matlab or opencv to detect points, that describe the shapes.
but so far I could not find any code that fits my task.
Does anyone of you have created such a code I could use for my task or any other idea?
You can see a lot of hexagonal patches in different depths also.
the little spots with an greater pixel value are the wanted particles!
Image processing is quite an involved area so there are no hard and fast rules.
But if it was me I would 'Mask' the image. This involves either defining what you want to keep or remove as a pixel 'Mask'. You then scan the mask over the image recursively and compare the mask to the image portion selected. You then select or remove the section (depending on your method) if it meets your criterion.
One such example of a criteria would be the spatial and grey-scale error weighted against a likelihood function (eg Chi-squared, square mean error etc.) or a Normal distribution that you define the uncertainty..
Some food for thought
Maybe you can try with the Hough transform:
https://en.wikipedia.org/wiki/Hough_transform
Matlab have an built-in function, hough, wich implements this, but only works for lines. Maybe you can start from that and change it to recognize hexagons.
Related
I am trying to figure out how SURF feature detection works. I think I have made some progress. I would like to know how off I am from what's really going on.
A template image you have already got stored and a real-world image
are compared on the basis of "key points" or some important features
in the two images.
The smallest Euclidean distance between the same points constitutes a
good match.
What constitutes an important feature or keypoint? A corner
(intersection of edges) or a blob (sharp change in intensity).
SURF uses blobs.
It uses a Hessian matrix for blob detection or feature extraction.
The Hessian matrix is a matrix of second derivatives: this is to
figure out the minima and maxima associated with the intensity of a
given region in the image.
sift/surf etc have 3 stages:
find features/keypoints that are likely to be found in different images of same object again (surf uses box filters afair). those features should be scale and rotation invariant if possible. corners, blobs etc are good and most often searched in multiple scales.
find the right "orientation" of that point so that if the image is rotated according to that orientation, both images are aligned in regard to that single keypoint.
computation of a "descriptor" that has information of how the neighborhood of the keypoint looks like (after orientation) in the right scale.
now your euclidean distance computation is done only on the descriptors, not on the keypoint locations!
it is important to know that step 1 isnt fixed for SURF. SURF in fact is step 2-3 but the authors give a suggestion how step 1 can be done to have some synergies with steps 2-3. the synergy is that both, step 1 and 3 use integral images to speed things up, so the integral image has to be computed only once.
I have been working a self project in image processing and robotics where instead robot as usual detecting colors and picking out the object, it tries to detect the holes(resembling different polygons) on the board. For a better understanding of the setup here is an image:
As you can see I have to detect these holes, find out their shapes and then use the robot to fit the object into the holes. I am using a kinect depth camera to get the depth image. The pic is shown below:
I was lost in thought of how to detect the holes with the camera, initially using masking to remove the background portion and some of the foreground portion based on the depth measurement,but this did not work out as, at different orientations of the camera the holes would merge with the board... something like inranging (it fully becomes white). Then I came across adaptiveThreshold function
adaptiveThreshold(depth1,depth3,255,ADAPTIVE_THRESH_GAUSSIAN_C,THRESH_BINARY,7,-1.0);
With noise removal using erode, dilate, and gaussian blur; which detected the holes in a better manner as shown in the picture below. Then I used the cvCanny edge detector to get the edges but so far it has not been good as shown in the picture below.After this I tried out various feature detectors from SIFT, SURF, ORB, GoodFeaturesToTrack and found out that ORB gave the best times and the features detected. After this I tried to get the relative camera pose of a query image by finding its keypoints and matching those keypoints for good matches to be given to the findHomography function. The results are as shown below as in the diagram:
In the end i want to get the relative camera pose between the two images and move the robot to that position using the rotational and translational vectors got from the solvePnP function.
So is there any other method by which I could improve the quality of the
holes detected for the keypoints detection and matching?
I had also tried contour detection and approxPolyDP but the approximated shapes are not really good:
I have tried tweaking the input parameters for the threshold and canny functions but
this is the best I can get
Also ,is my approach to get the camera pose correct?
UPDATE : No matter what I tried I could not get good repeatable features to map. Then I read online that a depth image is cheap in resolution and its only used for stuff like masking and getting the distances. So , it hit me that the features are not proper because of the low resolution image with its messy edges. So I thought of detecting features on a RGB image and using the depth image to get only the distances of those features. The quality of features I got were literally off the charts.It even detected the screws on the board!! Here are the keypoints detected using GoodFeaturesToTrack keypoint detection..
I met an another hurdle while getting the distancewith the distances of the points not coming out properly. I searched for possible causes and it occured to me after quite a while that there was a offset in the RGB and depth images because of the offset between the cameras.You can see this from the first two images. I then searched the net on how to compensate this offset but could not find a working solution.
If anyone one of you could help me in compensate the offset,it would be great!
UPDATE: I could not make good use of the goodFeaturesToTrack function. The function gives the corners in Point2f type .If you want to compute the descriptors we need the keypoints and converting Point2f to Keypoint with the code snippet below leads to the loss of scale and rotational invariance.
for( size_t i = 0; i < corners1.size(); i++ )
{
keypoints_1.push_back(KeyPoint(corners1[i], 1.f));
}
The hideous result from the feature matching is shown below .
I have to start on different feature matchings now.I'll post further updates. It would be really helpful if anyone could help in removing the offset problem.
Compensating the difference between image output and the world coordinates:
You should use good old camera calibration approach for calibrating the camera response and possibly generating a correction matrix for the camera output (in order to convert them into real scales).
It's not that complicated once you have printed out a checkerboard template and capture various shots. (For this application you don't need to worry about rotation invariance. Just calibrate the world view with the image array.)
You can find more information here: http://www.vision.caltech.edu/bouguetj/calib_doc/htmls/own_calib.html
--
Now since I can't seem to comment on the question, I'd like to ask if your specific application requires the machine to "find out" the shape of the hole on the fly. If there are finite amount of hole shapes, you may then model them mathematically and look for the pixels that support the predefined models on the B/W edge image.
Such as (x)^2+(y)^2-r^2=0 for a circle with radius r, whereas x and y are the pixel coordinates.
That being said, I believe more clarification is needed regarding the requirements of the application (shape detection).
If you're going to detect specific shapes such as the ones in your provided image, then you're better off using a classifer. Delve into Haar classifiers, or better still, look into Bag of Words.
Using BoW, you'll need to train a bunch of datasets, consisting of positive and negative samples. Positive samples will contain N unique samples of each shape you want to detect. It's better if N would be > 10, best if >100 and highly variant and unique, for good robust classifier training.
Negative samples would (obviously), contain stuff that do not represent your shapes in any way. It's just for checking the accuracy of the classifier.
Also, once you have your classifier trained, you could distribute your classifier data (say, suppose you use SVM).
Here are some links to get you started with Bag of Words:
https://gilscvblog.wordpress.com/2013/08/23/bag-of-words-models-for-visual-categorization/
Sample code:
http://answers.opencv.org/question/43237/pyopencv_from-and-pyopencv_to-for-keypoint-class/
I want to detect some elements in an Image.
For this goal, i get the image and the specified element (like a nose) and from Pixel(0,0) start to search for my element.
But the software performance is awful because i traverse the pixels one by one.
I think i need some smart algorithm for this problem.
And maybe the machine learning algorithm useful for this.
What's your idea?
I would start with viola jones object detection framework.
This is a supervised learning technique, that allows you to detect any kind of object with high provavility.
(even though the article mainly refers to faces, but it is designed for general objects..).
If you chose this approach - your main chore is going to be to obtain a classified training set. You can later evaluate how good your algorithm is using cross-validation.
AFAIK, it is implemented in OpenCV library (I am not familiar with the library to offer help)
You can do a very fast cross correlation using the Fourier transformation of your image and search pattern
A good implementation is for example OpenCV's matchTemplate function
This will work best if your pattern always has the same rotation and scale accross your image.
If it does not, you can repeat the search with several scaled/rotated versions of your pattern.
One advantage of this approach is that no training phase is required.
Another, simpler approach that would work in particular with your pattern is this:
Use connected component labeling to identify blobs with the right number of white pixels to be the center rectangle of your element. This will eliminate all but a few false positives. Concentrate your search on the remaining few spots.
Again OpenCV has a nice Blob library for that sort of stuff.
If you're looking for simple geometric shapes in computer-generated images like the example you provided, then you don't need to bother with machine learning.
For example, here's one of the components you're trying to find in the original image:
(Image removed by request)
Assuming this component is always drawn at the same dimensions, the top and bottom lines are always going to be 21 pixels apart. You can narrow down your search space considerably by combining this image with a copy of itself shifted vertically by 21 pixels, and taking the lighter of the two images as the pixel value at each position.
(Image removed by request)
Similarly, the vertical lines at the left and right of this component are 47 pixels apart, so we can repeat this process with a 47px horizontal shift. This results in a vertical bar about 24px tall at the position of the component.
(Image removed by request)
You can detect these bars quite easily by looking for runs of black pixels between 22 and 26 pixels long in the vertical columns of the processed image. This will provide you with a short list of candidate positions where you can check for the presence of this component more thoroughly, e.g. by calculating a local 2D cross correlation.
Here are the results after processing the whole image. Reaching this stage should only take a few milliseconds.
(Image removed by request)
I have a query on calculation of best matching point of one image to another image through intensity based registration. I'd like to have some comments on my algorithm:
Compute the warp matrix at this iteration
For every point of the image A,
2a. We warp the particular image A pixel coordinates with the warp matrix to image B
2b. Perform interpolation to get the corresponding intensity form image B if warped point coordinate is in image B.
2c. Calculate the similarity measure value between warped pixel A intensity and warped image B intensity
Cycle through every pixel in image A
Cycle through every possible rotation and translation
Would this be okay? Is there any relevant opencv code we can reference?
Comments on algorithm
Your algorithm appears good although you will have to be careful about:
Edge effects: You need to make sure that the algorithm does not favour matches where most of image A does not overlap image B. e.g. you may wish to compute the average similarity measure and constrain the transformation to make sure that at least 50% of pixels overlap.
Computational complexity. There may be a lot of possible translations and rotations to consider and this algorithm may be too slow in practice.
Type of warp. Depending on your application you may also need to consider perspective/lighting changes as well as translation and rotation.
Acceleration
A similar algorithm is commonly used in video encoders, although most will ignore rotations/perspective changes and just search for translations.
One approach that is quite commonly used is to do a gradient search for the best match. In other words, try tweaking the translation/rotation in a few different ways (e.g. left/right/up/down by 16 pixels) and pick the best match as your new starting point. Then repeat this process several times.
Once you are unable to improve the match, reduce the size of your tweaks and try again.
Alternative algorithms
Depending on your application you may want to consider some alternative methods:
Stereo matching. If your 2 images come from stereo camera then you only really need to search in one direction (and OpenCV provides useful methods to do this)
Known patterns. If you are able to place a known pattern (e.g. a chessboard) in both your images then it becomes a lot easier to register them (and OpenCV provides methods to find and register certain types of pattern)
Feature point matching. A common approach to image registration is to search for distinctive points (e.g. types of corner or more general places of interest) and then try to find matching distinctive points in the two images. For example, OpenCV contains functions to detect SURF features. Google has published a great paper on using this kind of approach in order to remove rolling shutter noise that I recommend reading.
My aim is to detect the vein pattern in leaves which characterize various species of plants
I have already done the following:
Original image:
After Adaptive thresholding:
However the veins aren't that clear and get distorted , Is there any way i could get a better output
EDIT:
I tried color thresholding my results are still unsatisfactory i get the following image
Please help
The fact that its a JPEG image is going to give the "block" artifacts, which in the example you posted causes most square areas around the veins to have lots of noise, so ideally work on an image that's not been through lossy compression. If that's not possible then try filtering the image to remove some of the noise.
The veins you are wanting to extract have a different colour from the background, leaf and shadow so some sort of colour based threshold might be a good idea. There was a recent S.O. question with some code that might help here.
After that some sort of adaptive normalisation would help increase the contrast before you threshold it.
[edit]
Maybe thresholding isn't an intermediate step that you want to do. I made the following by filtering to remove jpeg artifacts, doing some CMYK channel math (more cyan and black) then applying adaptive equalisation. I'm pretty sure you could then go on to produce (subpixel maybe) edge points using image gradients and non-maxima supression, and maybe use the brightness at each point and the properties of the vein structure (mostly joining at a tangent) to join the points into lines.
In the past I made good experiences with the Edge detecting algorithm difference of Gaussian. Which basically works like this:
You blur the image twice with the gaussian blurr algorithm but with differenct blur radii.
Then you calculate the difference between both images.
Pixel with same color beneath each other will creating a same blured color.
Pixel with different colors beneath each other wil reate a gradient which is depending on the blur radius. For bigger radius the gradient will stretch more far. For smaller ones it wont.
So basically this is bandpass filter. If the selected radii are to small a vain vill create 2 "parallel" lines. But since the veins of leaves are small compared with the extends of the Image you mostly find radii, where a vein results in 1 line.
Here I added th processed picture.
Steps I did on this picture:
desaturate (grayscaled)
difference of Gaussian. Here I blured the first Image with a radius of 10px and the second image with a radius of 2px. The result you can see below.
This is only a quickly created result. I would guess that by optimizing the parametes, you can even get better ones.
This sounds like something I did back in college with neural networks. The neural network stuff is a bit hard so I won't go there. Anyways, patterns are perfect candidates for the 2D Fourier transform! Here is a possible scheme:
You have training data and input data
Your data is represented as a the 2D Fourier transform
If your database is large you should run PCA on the transform results to convert a 2D spectrogram to a 1D spectrogram
Compare the hamming distance by testing the spectrum (after PCA) of 1 image with all of the images in your dataset.
You should expect ~70% recognition with such primitive methods as long as the images are of approximately the same rotation. If the images are not of the same rotation.you may have to use SIFT. To get better recognition you will need more intelligent training sets such as a Hidden Markov Model or a neural net. The truth is to getting good results for this kind of problem may be quite a lot of work.
Check out: https://theiszm.wordpress.com/2010/07/20/7-properties-of-the-2d-fourier-transform/